Gene polymorphisms and gene scores linked to low serum carotenoid status and their associations with metabolic disturbance and depressive symptoms in African-American adults

Abstract

Gene polymorphisms provide means to obtain unconfounded associations between carotenoids and various health outcomes. We tested whether gene polymophorisms and gene scores linked to serum carotenoid status are related to metabolic disturbance and depressive symptoms in African-American adults residing in Baltimore city, MD, using cross-sectional data from the Healthy Aging in Neighborhood of Diversity Across the Lifespan (HANDLS) study (Age range:30–64y, N=873–994). We examined 24 single nucleotide polymorphisms of various gene loci that were previously shown to be associated with low serum carotenoid status (SNP_lcar). Genetic risk scores (5 low specific-carotenoid risk scores (LSCRS: α-carotene, β-carotene, lutein+zeaxanthin, β-cryptoxanthin, lycopene) and 1 low total-carotenoid risk score (LTCRS: total carotenoids)) were created. SNP_lcar, LSCRS and LTCRS were entered as predictors for a number of health outcomes. Those included obesity, National Cholesterol Education Program (NCEP) Adult Treatment Panel (ATP) III metabolic syndrome (MetS) and its components, elevated homeostatic model assessment, Insulin Resistance (HOMA-IR), C-reactive protein (CRP), hyperuricemia and elevated depressive symptoms (EDS, Center for Epidemiologic Studies-Depression (CES-D) score≥16). Among key findings, SNP_lcar were not associated with the main outcomes after correction for multiple testing. However, an inverse association was found between LTCRS and HDL-C dyslipidemia. Specifically, the α-carotene and β-cryptoxanthin LSCRS were associated with lower odds of HDL-C dyslipidemia. However, the β-cryptoxanthin LSCRS was linked to a higher odds of EDS, with a linear dose-response relationship. In sum, gene risk scores linked to low serum carotenoids had mixed effects on HDL-C dyslipidemia and EDS. Further studies using larger African-American samples are needed.

Introduction

Oxidative stress, an imbalance between the production of reactive oxygen species (ROS) and the cell’s ability to scavenge those species with various antioxidants, has been implicated in the pathogenesis of many chronic diseases, including type 2 diabetes mellitus, cardiovascular disease, rheumatologic disorders and carcinogenesis.⁽¹⁾ Potential beneficial effects were recently ascribed to naturally occurring phytochemicals known as carotenoids which may reduce oxidative stress triggered by injury characterizing pathogenesis of those chronic diseases.⁽¹⁾ Although the primary dietary sources of carotenoids are fruits and vegetables, they are also found in bread, eggs, beverages (e.g. carrot and tomato juices), fats, and oils.⁽²⁾ Among more than 40 carotenoids in the human diet, only α-carotene, β-carotene, β-cryptoxanthin, lycopene, lutein+zeaxanthin (often combined together) are five carotenoids or groups of carotenoids that are consistently measurable in human serum.⁽²⁾

Some observational studies showed inverse associations between carotenoids and cardiovascular disease⁽³⁾, type 2 diabetes,^{(4; 5; 6; 7; 8)} and MetS in recent national surveys. ^{(9; 10; 11)} Moreover, in two recent studies, using NHANES and InChianti data, serum total carotenoid level was consistently inversely related to depressive symptoms.^{(12; 13)} However, findings were inconsistent in other studies.^{(14; 15; 16; 17)} It is also worth noting that obesity and its related disorders was associated with increased level of depressive symptoms in a number of studies.(e.g. ^{(18; 19; 20)}), suggesting co-morbidity between those conditions.

Importantly, it is unclear whether observed inverse relationships between serum carotenoids and MetS and/or depression are due to variations in carotenoids or determined by other carotenoid-containing food constituents. To identify an unconfounded role of carotenoids in health and disease, surrogate measures such as genetic polymorphisms have been used in recent studies. In fact, genome-wide association studies (GWAS) and candidate gene studies have uncovered genetic polymorphisms in a number of genes that were significantly associated with serum carotenoid status. Genes carrying the specific single nucleotide polymorphisms (SNP) that were commonly tested in the literature against serum carotenoid concentrations, were either directly (e.g. β,β-carotene 15,15′ mono-oxygenase, BCMO1) or indirectly (e.g. Apolipoprotein E, ApoE) related to carotenoid metabolism.^{(21; 22; 23; 24; 25; 26; 27; 28; 29; 30)} Many of these GWAS and candidate gene studies were conducted among individuals of European descent.

The present study’s overall aim was to assess whether genetic polymorphisms involved in carotenoid absorption, intracellular trafficking, and plasma transport are also related to a higher burden of metabolic disturbance and depressive symptoms. Our current study is focused on African-American adults from the Healthy Aging in Neighborhoods of Diversity across the Life Span Study (HANDLS) cohort providing the first opportunity to examine those relationships within that racial/ethnic group. The findings could elucidate whether metabolic disturbance and/or depression are associated with genetic polymorphisms that are in turn related to low serum carotenoid status.

Materials and Methods

Database and study population

Initiated in 2004 as an ongoing prospective cohort study, the HANDLS study used area probability sampling to recruit a socioeconomically diverse and representative sample of African Americans and whites (30–64 years old) living in Baltimore, Maryland.⁽³¹⁾ The HANDLS protocol was approved by the Institutional Review Board of the National Institute on Aging. Our study uses cross-sectional data from the baseline HANDLS cohort.

Three-thousand seven-hundred and twenty selected subjects participated in the household survey at phase 1 (Sample 1). Out of those, 2,436 (65.4%) had complete baseline phase 2 examinations (Sample 2). However, our data uses a subset with complete genetic data on a sample of African-American participants of HANDLS (n=1,024), (Sample 3). Of those, 873 had complete depressive symptoms data (Sample 4a) and 910–961 had complete data on metabolic outcomes (Sample 5a–5i).

Genetic data

Blood samples were collected for DNA extraction, and genome-wide genotyping was completed for 1,024 HANDLS participants using Illumina 1M SNP coverage. Online Supplemental Material 1(OSM1) describes the methods used further.

Selection of SNP of interest for our analyses was solely based on those detected in previous GWAS and candidate gene studies as highly significant predictors of serum carotenoid status.^{(22; 28; 32)} These SNPs were extracted from high quality imputed genotypes. Most of those selected SNP were available in our database, with the exception of two BCDO2 SNP (W80X: bovine SNP; c.196C>T: sheep SNP), which were not human SNP and one SCARB1 SNP (SR-BI: intron 5).⁽²²⁾ Other SNP (n=5) that were selected from one study,⁽²⁸⁾ were dropped for various reasons, the most common of which was high linkage disequilibrium with other selected SNP. None of the remaining SNPs were in strong LD with each others. Consequently, 24 distinctive SNP with reliable values were chosen. OSM2 provides a detailed description of those SNP.

SNP_lcar, LSCRS and LTCRS

From these 24 SNP, combinations were created that would allow assessing the effect of an increasing genetic risk of lower carotenoid level on binary measures of metabolic disturbance and depression. First, we examined independent effects of each SNP allele dosage that was previously shown to be associated with lower specific carotenoid or a group of carotenoids. To this end, from the 24 SNP, 24 genetic exposure variables were created and were termed SNP_lcar. SNP dosage was coded as is or reverse coded (0,1,2 or 2,1,0) depending on whether the minor allele was associated with lower carotenoid status or vice versa (See details in OSM2).

Moreover, to assess the collective associations of SNP linked to lower levels of specific and total carotenoid with the outcomes of interest, two risk scores were created: (1) LSCRS (Lower specific carotenoid risk score), by summing SNP_lcar values together that pertained to that specific carotenoid. In this computation, a SNPl_car was entered into an LSCRS, when previously shown to have the most significant association with a specific carotenoid (smallest p-value), particularly when multiple carotenoids were affected by the same SNP_lcar; (2) LTCRS (Lower total carotenoid risk score) by summing all SNP_lcar values together, reflecting low levels of all carotenoids. (OSM3 and OSM4) We assumed that each SNP_lcar was associated with levels of specific carotenoids based on previous findings in whites, despite potential ancestral differences in African-Americans, particularly in terms of linkage disequilibrium (LD) patterns.⁽³³⁾ Since a direct way to estimate effect size of each SNP_lcar on serum carotenoids was not available for African-Americans, we did not apply SNP-specific weights from previous studies on Whites to account for SNP-specific differences in carotenoid status effects. Thus, we simply summed the risk alleles or combinations of risk alleles together to obtain LSCRS and LTCRS, as was done in a previous study.⁽³⁴⁾ In each LSCRS, SNP_lcar included were specific to that particular carotenoid and were not double-counted in another LSCRS.

Anthropometric indices

Body weight and standing height were directly measured. Body Mass Index (BMI=weight/height², kg/m²) was calculated for each participant. Waist circumference (WC, in cm.) was measured using a tape measure starting from the hip bone and wrapping around the waist at the level of the navel. Obesity was defined as BMI≥30 kg.m⁻² while central obesity was defined as a component of MetS (See below).

Metabolic outcome variables

Systolic and diastolic blood pressure

The average of right and left sitting blood pressure values was taken to represent each of systolic and diastolic blood pressure levels (SBP and DBP, respectively) for our present analyses. Blood pressure was measured non-invasively using the brachial artery auscultation method with an aneroid manometer, a stethoscope, and an inflatable cuff.

Other metabolic risk factors

Following an overnight fast, blood samples were drawn from an antecubital vein. Total cholesterol, high density lipoprotein-cholesterol (HDL-C), triglycerides (TG), uric acid and glucose were assessed using a spectrophotometer (Olympus 5400). Fasting serum insulin was analyzed with a standard immunoassay test (Siemens/DPC Immulite 2000), and CRP was analyzed with an immunoturbidimeter (Siemens/Behring Nephelometer II).

Homeostatic model assessment, Insulin Resistance (HOMA-IR)⁽³⁵⁾ was computed with a cutoff point of 2.61 reflecting high insulin resistance level as suggested elsewhere.⁽³⁶⁾ Cutoffs for hyperuricemia were >7mg/dl in men and >6 mg/dl in women,⁽³⁷⁾ while elevated CRP was defined as >2.11 mg/L.⁽³⁸⁾

Metabolic syndrome (MetS)

Central obesity was defined by WC ≥ 102 cm or 40 inches (men), ≥ 88 cm or 35 inches (women).⁽³⁹⁾ This is one of five components of the National Cholesterol Education Program (NCEP) Adult Treatment Panel III (2005) (NCEP ATP III) main definition of MetS.⁽⁴⁰⁾ Using this definition, MetS was positive when three or more of the following criteria screened as positive: (1) WC>102 cm. for men and >88 cm. for women (2) SBP/DBP ≥ 130/85 mm Hg.; (3) Fasting glucose ≥ 100 mg/dL; (4) TG ≥ 150 mg/dL; (5) HDL-C<40 mg/dL for men and <50mg/dL for women.

Depressive symptoms assessment

Extensively trained psychometricians administered among others a baseline battery of cognitive and neuropsychological tests⁽⁴¹⁾ which included baseline depressive symptoms using the Center for Epidemiologic Studies Depression (CES-D) scale, a 20-item self-report symptom rating scale that emphasizes the affective, depressed mood component.⁽⁴²⁾ The invariant factor structure of the CES-D was recently shown using confirmatory factor analysis comparing NHANES I and HANDLS data.⁽⁴³⁾ A cut-point of 16 was used to assess EDS in all analyses.

Covariates

Covariates considered as potential confounders included sex, age, education (<High School (grades 1–8), High school (grades 9–12), >High school (13+)), poverty income ratio (below vs. at or above the poverty line), smoking status (current smoker vs. not), drug use (current vs. past or never), and 10 principal components to control for any residual effects of population structure (See OSM1).

Statistical methods

Using Stata 13.0,⁽⁴⁴⁾ differences in means and associations of categorical variables across “genetic data completeness” were tested with t- and χ² tests, respectively. Second, multiple logistic regression models were conducted to test associations of SNP_lcar (entered separately in each model), five LCSRS (entered simultaneously) and one LTCRS with 10 binary outcomes (obesity, MetS(and 5 components), elevated HOMA-IR, elevated CRP, hyperuricemia and EDS. Adjusted odds ratios and 95%CI were estimated. Type I error was initially set at 0.05, with regression coefficients being assessed using the Wald test. Finally, to test linear dose-response relationships, quartiles of LSCRS and LTCRS were entered into the regression models as ordinal variables, and p-values for trend were computed from the Wald test. Additionally, non-linear associations were tested for each quartile compared with the lowest quartile (Q₁) as the common referent category. The SNP_lcar analyses were corrected for multiplte testing by setting type I error was reduced to α/k (k=24 is the number of SNP tested for each phenotype). Thus, the 2-sided p-values were presented uncorrected with significance level set at 0.05/24=0.002.

Results

According to Table 1, selected participants with complete data were generally older, but had less missing data on most socio-demographic and lifestyle variables compared to those without genetic data. All LSCRS (in their continuous form) were weakly to moderately correlated (R=−0.50 for lutein+zeaxanthin vs. β-cryptoxanthin to +0.044 for lutein+zeaxanthin vs. lycopene). Thus, it was possible to covary those gene scores in multiple logistic regression models. For descriptive purposes, mean dietary intakes of carotenoids (in μg/1,000kcal/d) are presented in Supplemental Figure 2, stratified by sex and PIR categories. Comparisons were made between categories and key findings included a higher intake among women of β-carotene and lutein+zeaxanthin. However, LTCRS was not correlated with total carotenoid intake in μg/1,000kcal/d (R=−0.03, P=0.35); (Supplemental Figure 3).

Table 1.

Participant characteristics¹ by genetic data completness; HANDLS study

	All African-American HANDLS participants		African-American HANDLS participants with genetic data		African-American HANDLS participants without genetic data
	N=2,198	Mean±SD or (%)	N=1,024	Mean±SD or (%)	N=1,174	Mean±SD or (%)
Characteristic
Socio-demographic, lifestyle factors
Age, y	2,198	47.7±9.3	1,024	48.5±9.0	1,174	47.0±9.5*
Female, %	1,200	(54.6)	569	(55.6)	631	(53.7)
Marital status, %	2,198		1,024		1,174
Married	610	(27.8)	427	(41.7)	183	(15.7)*
Missing	994	(45.2)	230	(22.5)	764	(65.0)
Education, %	2,198		1,024		1,174
<HS	117	(5.3)	49	(4.8)	68	(5.8)*
HS	1,421	(64.7)	638	(62.3)	783	(66.7)
>HS	647	(29.4)	333	(32.5)	314	(26.8)
Missing	13	(0.6)	4	(0.4)	9	(0.8)
Poverty income ratio<125%, %	1,156	(52.6)	542	(52.9)	614	(52.3)
Current smoking status, %	2,198		1,024		1,174
Currently smoking	781	(35.5)	469	(45.8)	312	(26.6)*
Missing	659	(30.0)	88	(8.6)	571	(48.6)
Currently using illicit drugs, %	2,198		1,024		1,174
Used any type	806	(37.7)	513	(50.1)	293	(25.0)*
Missing	694	(31.6)	86	(8.4)	608	(51.8)
Depressive symptoms
CES-D score	1,319	11.6±8.0	873	11.7±8.1	446	11.4±7.6
CES-D score≥16, %	351	(26.6)	235	(26.9)	116	(26.0)
Metabolic outcomes
BMI, kg.m−2	1,657	29.9±7.9	994	29.9±8.0	663	30.0±7.8
Obese, %, BMI≥30	711	(42.9)	416	(41.8)	295	(44.5)
Waist circumference, cm	1,589	98.4±17.5	961	98.5±17.5	628	98.2±17.5
Centrally obese, %	892	(56.1)	543	(56.5)	349	(55.6)
SBP, mm Hg	1,614	121.5±20.4	977	122.2±20.7	637	120.4±20.1
DBP, mm Hg	1,614	73.0±12.7	977	73.3±12.8	637	72.6±12.5
Elevated blood pressure, %	570	(35.4)	364	(37.3)	206	(32.3)*
HDL-C, mg/dL	1,576	55.5±18.5	989	55.1±18.1	587	56.1±19.2
Dyslipidemia, HDL-C, %	479	(30.4)	304	(30.7)	175	(29.8)
TG, mg/dL	1,577	109.1±73.0	989	106.7±67.6	588	113.2±81.2
Dyslipidemia, TG, %	288	(18.3)	178	(18.0)	110	(18.7)
Fasting blood glucose, mg/dL	1,578	103.7±43.1	989	104.5±42.8	589	102.2±43.6
Hyperglycemia, %	496	(31.4)	325	(32.9)	171	(29.0)
NCEP ATP III metabolic disturbances	1,480	1.71±1.28	928	1.74±1.28	552	1.64±1.29
MetS NCEP ATP III, %	396	(26.8)	260	(28.0)	136	(24.6)
HOMA-IR	1,561	3.20±4.41	986	3.13±3.90	575	3.32±5.17
Elevated HOMA-IR, %	617	(39.5)	385	(39.1)	232	(40.3)
CRP, mg/L2	1,518	4.63±6.72	967	4.61±6.73	551	4.66±6.72
Elevated CRP, %	741	(48.8)	466	(48.2)	275	(49.9)
Uric acid, mg/dL	1,576	5.55±1.66	989	5.55±1.65	587	5.55±1.68
Hyperuricemia, %	385	(24.4)	239	(24.2)	146	(24.9)

Abbreviations: ABCG5=ATP-binding cassette, subfamily G, member 5; ApoA=Apolipoprotein A; ApoB=Apolipoprotein B; ApoE=Apolipoprotein E; ATP III=Adult Treatment Panel III; BCMO1=beta-carotene mono-oxygenase 1 enzyme; BCDO2=Beta-carotene di-oxygenase 2 enzyme; BMI=Body Mass Index; CD36=thrombospondin receptor; CES-D=Center for Epidemiologic Studies-Depression scale; CRP=C-reactive protein; DBP=Diastolic Blood Pressure; EDS=Elevated Depressive Symptoms; FABP2=Fatty Acid Binding Protein 2; GWAS=Genome wide association studies; HANDLS=Healthy Aging in Neighborhood of Diversity Across the Lifespan; HDL-C=High Density Lipoprotein-Cholesterol; LIPC=Hepatic Lipase; HOMA-IR= Homeostatic model assessment, Insulin Resistance; HWE=Hardy-Weinberg equilibrium; LD=Linkage disequilibrium; LDL-C=Low-density Lipoprotein-Cholesterol; LPL=Lipoprotein Lipase; LSCRS=Low Specific Carotenoid Risk Score; LTCRS=Low Total Carotenoid Risk Score; MDS=Multi-Dimensional Scaling; MetS=Metabolic Syndrome; NCEP=National Cholesterol Education Program; NHANES I=National Health and Nutrition Examination Surveys I; Q1=Lowest quartile; Q4=Uppermost quartile; SBP=Systolic Blood Pressure; SCARB1=Scavenger receptor class B member 1; SNP=Single Nucleotide Polymorphism; SNPlcar=Single Nucleotide Polymorphism for lower carotenoid status; TG=Triglycerides; WC=Waist circumference.

¹Values are percent or Mean±Standard deviation (SD).

²Outliers with values of CRP >50 (n=12) were removed from this sample.

^*P<0.05 for null hypothesis of no difference by genetic data completeness, t-test or χ² test.

Examining each of the 24 SNPlcar in a separate model as a predictor for each of the outcomes of interest, controlling for key potential confounders (Table 2 and OSM5), a few associations emerged that were against the hypothesized direction. Those included a putative inverse relation between SNPlcar_{17(BCMO1, β-carotene)} and obesity; SNPlcar_{2(APOB, β-carotene)} and several phenotypes indicative of inflammation (CRP), dyslipidemia (low HDL-C) and importantly NCEP ATP III MetS; SNPlcar_{10(BCMO1, β-cryptoxanthin)} and hypertension; SNPlcar_{19(CD36, lutein+zeaxanthin)} and dyslipidemia-TG and SNPlcar_{23(LPL, α-carotene)} with elevated HOMA-IR. On the other hand, a number of SNPlcar showed positive associations that were in line with the hypothesis, mainly within the BCMO1 locus. Those included SNPlcar14(BCMO1, β-cryptoxanthin) and EDS; SNPlcar12(BCMO1, α-carotene) and central obesity; SNPlcar14(BCMO1, β-carotene) and SNPlcar16(BCMO1, β-carotene) with hypertension. However, none of the key findings survived Bonferroni correction.

Table 2.

Gene single nucleotide polymorphisms related to low carotenoid status (SNP_lcar) and their associations with selected binary metabolic outcomes (obesity, NCEP ATP III MetS and elevated depressive symptoms (EDS) among African-American adults: multiple logistic regression analysis¹

	Gene	OR	95%CI	p-value
Obesity (N=990)
SNPlcar1: rs6720173:C/G (0,1,2)	ABCG5	0.86	(0.70;1.08)	0.20
SNPlcar2: rs934197:T/C(0,1,2)	ApoB	0.91	(0.66;1.25)	0.56
SNPlcar3: rs675:TT=1 vs. others=0	Apo A-IV	0.97	(0.70;1.33)	0.84
SNPlcar4: ApoE3/2 =2 vs. E3/3=0, others=1	ApoE	0.94	(0.77;1.14)	0.54
SNPlcar5: rs6564851:T/G(0,1,2)	BCMO1	1.10	(0.91;1.34)	0.32
SNPlcar6: rs6564851:T/G(2,1,0)	BCMO1	0.90	(0.75;1.10)	0.32
SNPlcar7: rs12934922: T/A (2,1,0)+ rs7501331: T/C(2,1,0)	BCMO1	1.04	(0.79;1.35)	0.79
SNPlcar8: rs7501331:T/C(2,1,0)	BCMO1	0.83	(0.50;1.37)	0.47
SNPlcar9: rs56389940:A/C(0,1,2)	BCMO1	0.84	(0.63;1.14)	0.27
SNPlcar10: rs12918164:A/G(0,1,2)	BCMO1	1.04	(0.61;1.76)	0.88
SNPlcar11: rs10048138:A/G(0, 1,2)	BCMO1	1.10	(0.89;1.35)	0.35
SNPlcar12: rs4889293:G/C(0,1,2)	BCMO1	1.20	(0.92;1.56)	0.17
SNPlcar13: rs12934922:T/A(0,1,2)	BCMO1	0.89	(0.67;1.20)	0.46
SNPlcar14: rs4448930:C/G(0,1,2)	BCMO1	0.83	(0.57;1.19)	0.31
SNPlcar15: rs1165428:A/G(2,1,0)	BCMO1	1.18	(0.88;1.60)	0.27
SNPlcar16: rs6420424:G/A(2,1,0)	BCMO1	1.20	(0.99;1.46)	0.06
SNPlcar17:rs8044334:G/T(0,1,2)	BCMO1	0.80	(0.67;0.98)	0.028
SNPlcar18:rs1761667:A/G(2,1,0)	CD36	0.92	(0.76;1.12)	0.40
SNPlcar19: rs13230419:T/C: CC=1 vs. others=0	CD36	1.03	(0.76;1.39)	0.85
SNPlcar20: rs1800588:T/C: TT=1 vs. others=0	LIPC	0.82	(0.60;1.11)	0.20
SNPlcar21: rs1800588:T/C(0,1,2)	LIPC	1.13	(0.93;1.37)	0.21
SNPlcar22: rs1799883: A/G: GG=1 vs. others=0	FABP2	3.41	(0.88;13.13)	0.07
SNPlcar23: rs328: G/C: (2,1,0)	LPL	0.75	(0.51;1.12)	0.16
SNPlcar24: rs61932577:A/G: GG=1 vs. others=0	SCARB1	2.40	(1.03;5.61)	0.043
NCEP ATP III MetS (N=928)
SNPlcar1: rs6720173:C/G (0,1,2)	ABCG5	0.97	(0.77;1.23)	0.81
SNPlcar2: rs934197:T/C(0,1,2)	ApoB	0.72	(0.52;1.00)	0.048
SNPlcar3: rs675:TT=1 vs. others=0	Apo A-IV	0.95	(0.67;1.35)	0.77
SNPlcar4: ApoE3/2 =2 vs. E3/3=0, others=1	ApoE	0.99	(0.80;1.23)	0.94
SNPlcar5: rs6564851:T/G(0,1,2)	BCMO1	0.97	(0.78;1.20)	0.76
SNPlcar6: rs6564851:T/G(2,1,0)	BCMO1	1.03	(0.83;1.28)	0.76
SNPlcar7: rs12934922: T/A (2,1,0)+ rs7501331: T/C(2,1,0)	BCMO1	0.95	(0.71;1.28)	0.75
SNPlcar8: rs7501331:T/C(2,1,0)	BCMO1	1.02	(0.60;1.74)	0.94
SNPlcar9: rs56389940:A/C(0,1,2)	BCMO1	0.88	(0.64;1.22)	0.47
SNPlcar10: rs12918164:A/G(0,1,2)	BCMO1	0.74	(0.43;1.29)	0.29
SNPlcar11: rs10048138:A/G(0, 1,2)	BCMO1	0.86	(0.69;1.08)	0.21
SNPlcar12: rs4889293:G/C(0,1,2)	BCMO1	1.07	(0.81;1.43)	0.62
SNPlcar13: rs12934922:T/A(0,1,2)	BCMO1	1.07	(0.77;1.49)	0.69
SNPlcar14: rs4448930:C/G(0,1,2)	BCMO1	1.09	(0.72;1.65)	0.69
SNPlcar15: rs1165428:A/G(2,1,0)	BCMO1	1.13	(0.81;1.57)	0.46
SNPlcar16: rs6420424:G/A(2,1,0)	BCMO1	1.15	(0.93;1.42)	0.20
SNPlcar17:rs8044334:G/T(0,1,2)	BCMO1	1.03	(0.84;1.27)	0.75
SNPlcar18:rs1761667:A/G(2,1,0)	CD36	0.88	(0.70;1.10)	0.27
SNPlcar19: rs13230419:T/C: CC=1 vs. others=0	CD36	0.93	(0.67;1.29)	0.66
SNPlcar20: rs1800588:T/C: TT=1 vs. others=0	LIPC	0.92	(0.66;1.29)	0.64
SNPlcar21: rs1800588:T/C(0,1,2)	LIPC	1.02	(0.83;1.27)	0.83
SNPlcar22: rs1799883: A/G: GG=1 vs. others=0	FABP2	0.39	(0.13;1.18)	0.10
SNPlcar23: rs328: G/C: (2,1,0)	LPL	0.58	(0.36;0.92)	0.022
SNPlcar24: rs61932577:A/G: GG=1 vs. others=0	SCARB1	0.55	(0.25;1.20)	0.13
Elevated depressive symptoms (EDS, CES-D≥16) (N=873)
SNPlcar1: rs6720173:C/G (0,1,2)	ABCG5	0.96	(0.75;1.23)	0.73
SNPlcar2: rs934197:T/C(0,1,2)	ApoB	0.86	(0.61;1.23)	0.41
SNPlcar3: rs675:TT=1 vs. others=0	Apo A-IV	1.30	(0.89;1.90)	0.18
SNPlcar4: ApoE3/2 =2 vs. E3/3=0, others=1	ApoE	0.94	(0.75;1.18)	0.60
SNPlcar5: rs6564851:T/G(0,1,2)	BCMO1	0.95	(0.76;1.18)	0.63
SNPlcar6: rs6564851:T/G(2,1,0)	BCMO1	1.06	(0.84;1.32)	0.63
SNPlcar7: rs12934922: T/A (2,1,0)+ rs7501331: T/C(2,1,0)	BCMO1	1.03	(0.77;1.38)	0.83
SNPlcar8: rs7501331:T/C(2,1,0)	BCMO1	1.14	(0.67;1.92)	0.64
SNPlcar9: rs56389940:A/C(0,1,2)	BCMO1	1.39	(1.00;1.94)	0.05
SNPlcar10: rs12918164:A/G(0,1,2)	BCMO1	1.49	(0.77;2.89)	0.24
SNPlcar11: rs10048138:A/G(0, 1,2)	BCMO1	1.02	(0.81;1.30)	0.83
SNPlcar12: rs4889293:G/C(0,1,2)	BCMO1	0.77	(0.58;1.03)	0.08
SNPlcar13: rs12934922:T/A(0,1,2)	BCMO1	1.00	(0.72;1.42)	0.96
SNPlcar14: rs4448930:C/G(0,1,2)	BCMO1	2.05	(1.27;3.31)	0.003
SNPlcar15: rs1165428:A/G(2,1,0)	BCMO1	0.72	(0.51;1.00)	0.05
SNPlcar16: rs6420424:G/A(2,1,0)	BCMO1	1.07	(0.86;1.34)	0.52
SNPlcar17:rs8044334:G/T(0,1,2)	BCMO1	0.95	(0.76;1.18)	0.64
SNPlcar18:rs1761667:A/G(2,1,0)	CD36	0.90	(0.71;1.14)	0.40
SNPlcar19: rs13230419:T/C: CC=1 vs. others=0	CD36	1.41	(0.99;2.00)	0.06
SNPlcar20: rs1800588:T/C: TT=1 vs. others=0	LIPC	1.04	(0.74;1.47)	0.81
SNPlcar21: rs1800588:T/C(0,1,2)	LIPC	0.93	(0.74;1.16)	0.54
SNPlcar22: rs1799883: A/G: GG=1 vs. others=0	FABP2	2.44	(0.30;20.1)	0.41
SNPlcar23: rs328: G/C: (2,1,0)	LPL	1.12	(0.74;1.71)	0.60
SNPlcar24: rs61932577:A/G: GG=1 vs. others=0	SCARB1	1.72	(0.60;4.97)	0.32

Abbreviations: ABCG5=ATP-binding cassette, subfamily G, member 5; ApoA=Apolipoprotein A; ApoB=Apolipoprotein B; ApoE=Apolipoprotein E; ATP III=Adult Treatment Panel III; BCMO1=beta-carotene mono-oxygenase 1 enzyme; BCDO2=Beta-carotene di-oxygenase 2 enzyme; BMI=Body Mass Index; CD36=thrombospondin receptor; CES-D=Center for Epidemiologic Studies-Depression scale; CRP=C-reactive protein; DBP=Diastolic Blood Pressure; EDS=Elevated Depressive Symptoms; FABP2=Fatty Acid Binding Protein 2; GWAS=Genome wide association studies; HANDLS=Healthy Aging in Neighborhood of Diversity Across the Lifespan; HDL-C=High Density Lipoprotein-Cholesterol; LIPC=Hepatic Lipase; HOMA-IR= Homeostatic model assessment, Insulin Resistance; HWE=Hardy-Weinberg equilibrium; LD=Linkage disequilibrium; LDL-C=Low-density Lipoprotein-Cholesterol; LPL=Lipoprotein Lipase; LSCRS=Low Specific Carotenoid Risk Score; LTCRS=Low Total Carotenoid Risk Score; MDS=Multi-Dimensional Scaling; MetS=Metabolic Syndrome; NCEP=National Cholesterol Education Program; NHANES I=National Health and Nutrition Examination Surveys I; Q1=Lowest quartile; Q4=Uppermost quartile; SBP=Systolic Blood Pressure; SCARB1=Scavenger receptor class B member 1; SNP=Single Nucleotide Polymorphism; SNPlcar=Single Nucleotide Polymorphism for lower carotenoid status; TG=Triglycerides; WC=Waist circumference.

¹Each SNP_lcar was entered in a separate multiple logistic regression model as the main predictor. SNP allele dosage was coded as is or reverse coded (0,1,2 or 2,1,0) depending on whether the minor allele was associated with lower carotenoid status or vice versa versa (See details in OSM2). Covariate entered as potential confounder were: sex, age, poverty income ratio (<125% vs. ≥125%), education (<HS, HS, >HS), marital status (current, former or never, missing), smoking status (current, former or never, missing), drug use (current, past or never, missing) and 10 principal components to adjust for population structure.

When combining SNPlcar into genetic risk scores reflecting lower levels of specific carotenoids (i.e. LSCRS) and examining their associations with mutliple outcomes (Table 3), several findings emerged. First, the α-carotene LSCRS was associated with a lower odds of HDL-C dyslipidemia (Q₄ vs. Q₁: OR=0.65, 95% CI: 0.44–0.97; p=0.037, p-trend=0.045), with a similar pattern observed for the β-cryptoxanthin LSCRS (Q₄ vs. Q₁: OR=0.61; 95%CI:0.38–0.96; p=0.033; p-trend=0.039). In contrast, that same LSCRS was associated with a higher odds of EDS (Q₄ vs. Q₁: OR=1.83; 95%CI:1.07–3.12, p=0.026; p-trend=0.047).

Table 3.

Low specific carotenoid risk scores (LSCRS, quartiles) and their associations with selected binary metabolic outcomes (central obesity, NCEP ATP III MetS and its components, elevated HOMA-IR, elevated CRP and hyperuriciemia) and elevated depressive symptoms (EDS) among African-American adults: multiple logistic regression analysis¹

	Q1	Q2			Q3			Q4
		OR	95%CI	p-value	OR	95%CI	p-value	OR	95%CI	p-value
Obesity (N=990)
α-carotene (P-trend=0.60)	1	0.93	(0.63;1.37)	0.70	0.93	(0.63;1.36)	0.70	0.87	(0.59;1.28)	0.49
β-carotene(P-trend=0.92)	1	0.82	(0.55;1.23)	0.34	1.06	(0.72;1.56)	0.77	0.88	(0.59;1.32)	0.55
Lutein+zeaxanthin(P-trend=0.59)	1	0.88	(0.59;1.31)	0.53	0.92	(0.61;1.41)	0.72	0.85	(0.55;1.31)	0.46
β-cryptoxanthin(P-trend=0.44)	1	1.00	(0.66;1.47)	0.96	0.80	(0.53;1.22)	0.30	0.90	(0.58;1.40)	0.65
Lycopene(P-trend=0.95)	1	1.99	(0.86;4.61)	0.11	1.32	(0.73;2.36)	0.46	1.25	(0.70;2.22)	0.46
Central obesity(N=957)
α-carotene (P-trend=0.79)	1	1.09	(0.71;1.68)	0.68	1.34	(0.87;2.06)	0.19	0.97	(0.64;1.49)	0.90
β-carotene (P-trend=0.50)	1	0.80	(0.51;1.24)	0.32	1.39	(0.89;2.15)	0.14	1.01	(0.64;1.57)	0.97
Lutein+zeaxanthin(P-trend=0.37)	1	0.92	(0.59;1.44)	0.72	0.96	(0.60;1.54)	0.88	0.74	(0.45;1.21)	0.23
β-cryptoxanthin(P-trend=0.93)	1	0.92	(0.59;1.43)	0.72	1.19	(0.75;1.89)	0.46	0.97	(0.59;1.60)	0.92
Lycopene(P-trend=0.70)	1	2.81	(1.09;7.26)	0.033	1.80	(0.95;3.42)	0.07	1.36	(0.72;2.57)	0.34
NCEP ATP III MetS(N=928)
α-carotene (P-trend=0.38)	1	0.74	(0.48;1.14)	0.17	0.77	(0.50;1.18)	0.22	0.81	(0.53;1.23)	0.32
β-carotene (P-trend=0.30)	1	1.07	(0.68;1.69)	0.76	1.34	(0.87;2.06)	0.19	1.18	(0.75;1.86)	0.47
Lutein+zeaxanthin (P-trend=0.13)	1	0.58	(0.37;0.90)	0.014	0.52	(0.33;0.84)	0.007	0.69	(0.43;1.11)	0.13
β-cryptoxanthin (P-trend=0.67)	1	0.96	(0.62;1.50)	0.87	0.96	(0.60;1.53)	0.87	0.84	(0.51;1.38)	0.50
Lycopene (P-trend=0.06)	1	1.79	(0.74;4.35)	0.20	1.17	(0.63;2.19)	0.61	0.82	(0.51;1.38)	0.50
Elevated blood pressure(N=977)
α-carotene (P-trend=0.58)	1	1.36	(0.92;2.03)	0.12	1.05	(0.71;1.57)	0.79	0.97	(0.65;1.46)	0.90
β-carotene (P-trend=0.61)	1	1.30	(0.87;1.95)	0.21	1.13	(0.76;1.69)	0.55	1.21	(0.80;1.83)	0.37
Lutein+zeaxanthin (P-trend=0.30)	1	0.79	(0.53;1.19)	0.26	0.68	(0.44;1.04)	0.08	0.84	(0.54;1.31)	0.45
β-cryptoxanthin (P-trend=0.97)	1	1.06	(0.71;1.59)	0.78	0.96	(0.62;1.47)	0.86	0.74	(0.42;1.33)	0.32
Lycopene (P-trend=0.12)	1	0.90	(0.39;2.09)	0.82	0.96	(0.54;1.71)	0.90	0.75	(0.42;1.33)	0.32
Dyslipidemia, HDL-C(N=989)
α-carotene (P-trend=0.045)	1	0.71	(0.47;1.06)	0.09	0.68	(0.45;1.01)	0.06	0.65	(0.44;0.97)	0.037
β-carotene (P-trend=0.51)	1	0.91	(0.60;1.39)	0.68	1.23	(0.82;1.83)	0.32	1.02	(0.67;1.54)	0.93
Lutein+zeaxanthin (P-trend=0.34)	1	0.73	(0.48;1.10)	0.13	0.65	(0.42;1.00)	0.05	0.80	(0.51;1.25)	0.32
β-cryptoxanthin (P-trend=0.039)	1	0.82	(0.54;1.24)	0.34	0.70	(0.46;1.07)	0.10	0.61	(0.38;0.96)	0.033
Lycopene (P-trend=0.14)	1	1.06	(0.45;2.48)	0.89	0.89	(0.50;1.60)	0.70	0.75	(0.42;1.34)	0.33
Dyslipidemia, TG(N=989)
α-carotene (P-trend=0.48)	1	1.21	(0.76;1.91)	0.43	0.69	(0.42;1.15)	0.16	1.01	(0.63;1.62)	0.96
β-carotene (P-trend=0.71)	1	1.02	(0.62;1.69)	0.93	1.06	(0.65;1.73)	0.82	1.11	(0.68;1.83)	0.67
Lutein+zeaxanthin (P-trend=0.39)	1	0.51	(0.31;0.83)	0.007	0.68	(0.41;1.12)	0.13	0.80	(0.47;1.35)	0.40
β-cryptoxanthin (P-trend=0.63)	1	0.78	(0.47;1.28)	0.33	0.95	(0.57;1.57)	0.83	0.83	(0.48;1.44)	0.51
Lycopene (P-trend=0.52)	1	2.02	(0.72;5.65)	0.18	1.80	(0.84;3.89)	0.13	1.26	(0.59;2.75)	0.55
Hyperglycemia(N=989)
α-carotene (P-trend=0.84)	1	0.83	(0.55;1.24)	0.35	0.85	(0.57;1.27)	0.43	1.06	(0.71;1.57)	0.78
β-carotene (P-trend=0.68)	1	1.22	(0.82;1.83)	0.33	1.13	(0.76;1.69)	0.61	0.94	(0.62;1.42)	0.76
Lutein+zeaxanthin (P-trend=0.39)	1	0.80	(0.54;1.19)	0.28	0.65	(0.42;1.00)	0.05	0.88	(0.56;1.37)	0.56
β-cryptoxanthin (P-trend=0.61)	1	0.84	(0.56;1.26)	0.39	0.84	(0.55;1.28)	0.43	0.83	(0.53;1.30)	0.42
Lycopene (P-trend=0.92)	1	2.16	(0.91;5.13)	0.08	1.65	(0.87;3.06)	0.12	1.39	(0.75;2.57)	0.30
Elevated HOMA-IR(N=986)
α-carotene (P-trend=0.94)	1	0.87	(0.59;1.27)	0.47	0.81	(0.55;1.18)	0.27	1.00	(0.69;1.47)	0.97
β-carotene (P-trend=0.35)	1	0.91	(0.62;1.34)	0.62	1.07	(0.73;1.56)	0.75	0.78	(0.52;1.16)	0.21
Lutein+zeaxanthin (P-trend=0.85)	1	0.81	(0.55;1.19)	0.28	0.84	(0.56;1.27)	0.41	0.95	(0.62;1.46)	0.83
β-cryptoxanthin (P-trend=0.15)	1	1.19	(0.80;1.76)	0.40	1.35	(0.90;2.03)	0.15	1.33	(0.86;2.05)	0.21
Lycopene (P-trend=0.46)	1	1.91	(0.83;4.41)	0.13	1.65	(0.92;2.97)	0.10	1.56	(0.87;2.79)	0.14
Elevated CRP(N=963)
α-carotene (P-trend=0.92)	1	0.98	(0.67;1.43)	0.91	0.84	(0.57;1.23)	0.37	1.06	(0.72;1.55)	0.76
β-carotene (P-trend=0.72)	1	1.27	(0.86;1.87)	0.23	1.33	(0.91;1.95)	0.14	0.92	(0.62;1.36)	0.67
Lutein+zeaxanthin (P-trend=0.35)	1	0.95	(0.65;1.41)	0.81	0.99	(0.67;1.50)	0.97	0.81	(0.53;1.25)	0.35
β-cryptoxanthin (P-trend=0.84)	1	1.00	(0.68;1.48)	0.99	1.02	(0.68;1.53)	0.92	0.94	(0.61;1.44)	0.76
Lycopene (P-trend=0.12)	1	0.97	(0.42;2.25)	0.94	0.78	(0.44;1.36)	0.38	0.69	(0.39;1.20)	0.19
Hyperuricemia(N=989)
α-carotene (P-trend=0.51)	1	0.64	(0.41;0.98)	0.040	0.66	(0.43;1.02)	0.06	0.89	(0.58;1.34)	0.57
β-carotene (P-trend=0.46)	1	1.11	(0.71;1.74)	0.63	1.24	(0.80;1.93)	0.33	1.12	(0.71;1.77)	0.61
Lutein+zeaxanthin (P-trend=0.89)	1	0.79	(0.51;1.23)	0.30	1.01	(0.64;1.59)	0.98	0.91	(0.56;1.48)	0.72
β-cryptoxanthin (P-trend=0.44)	1	0.75	(0.48;1.17)	0.20	0.85	(0.54;1.33)	0.47	0.77	(0.47;1.25)	0.29
Lycopene (P-trend=0.50)	1	1.35	(0.53;3.44)	0.53	1.35	(0.71;2.56)	0.36	1.02	(0.53;1.94)	0.96
Elevated depressive symptoms (EDS, CES-D≥16) (N=873)
α-carotene (P-trend=0.31)	1	0.61	(0.38;0.96)	0.034	0.80	(0.50;1.23)	0.30	0.71	(0.46;1.11)	0.14
β-carotene (P-trend=0.65)	1	1.14	(0.71;1.82)	0.58	0.98	(0.62;1.54)	0.92	0.93	(0.58;1.48)	0.76
Lutein+zeaxanthin (P-trend=0.11)	1	0.98	(0.61;1.56)	0.89	1.04	(0.63;1.72)	0.89	1.52	(0.92;2.54)	0.10
β-cryptoxanthin (P-trend=0.047)	1	1.68	(1.05;2.70)	0.031	1.31	(0.80;2.14)	0.28	1.83	(1.07;3.12)	0.026
Lycopene (P-trend=0.37)	1	1.50	(0.60;3.75)	0.39	0.80	(0.41;1.57)	0.52	0.85	(0.44;1.66)	0.64

Abbreviations: ABCG5=ATP-binding cassette, subfamily G, member 5; ApoA=Apolipoprotein A; ApoB=Apolipoprotein B; ApoE=Apolipoprotein E; ATP III=Adult Treatment Panel III; BCMO1=beta-carotene mono-oxygenase 1 enzyme; BCDO2=Beta-carotene di-oxygenase 2 enzyme; BMI=Body Mass Index; CD36=thrombospondin receptor; CES-D=Center for Epidemiologic Studies-Depression scale; CRP=C-reactive protein; DBP=Diastolic Blood Pressure; EDS=Elevated Depressive Symptoms; FABP2=Fatty Acid Binding Protein 2; GWAS=Genome wide association studies; HANDLS=Healthy Aging in Neighborhood of Diversity Across the Lifespan; HDL-C=High Density Lipoprotein-Cholesterol; LIPC=Hepatic Lipase; HOMA-IR= Homeostatic model assessment, Insulin Resistance; HWE=Hardy-Weinberg equilibrium; LD=Linkage disequilibrium; LDL-C=Low-density Lipoprotein-Cholesterol; LPL=Lipoprotein Lipase; LSCRS=Low Specific Carotenoid Risk Score; LTCRS=Low Total Carotenoid Risk Score; MDS=Multi-Dimensional Scaling; MetS=Metabolic Syndrome; NCEP=National Cholesterol Education Program; NHANES I=National Health and Nutrition Examination Surveys I; Q1=Lowest quartile; Q4=Uppermost quartile; SBP=Systolic Blood Pressure; SCARB1=Scavenger receptor class B member 1; SNP=Single Nucleotide Polymorphism; SNPlcar=Single Nucleotide Polymorphism for lower carotenoid status; TG=Triglycerides; WC=Waist circumference.

¹All LSCRS were entered in the same multiple logistic regression model (as quartiles, with first quartile being the referent category) as main predictors, to assess their net association with each of the metabolic outcomes and with elevated depressive symptoms. Covariate entered as potential confounder were: sex, age, poverty income ratio (<125% vs. ≥125%), education (<HS, HS, >HS), marital status (current, former or never, missing), smoking status (current, former or never, missing), drug use (current, past or never, missing) and 10 principal components to adjust for population structure.

Moreover, a number of non-linear associations were also noted whereby a LSCRS was either inversely or positively associated with an outcome of interest when comparing one quartile with Q₁ but not others. For instance, a lower lutein+zeaxanthin gene risk score was associated with a lower odds of dyslipidmia-TA only when comparing Q₂ to Q₁ (OR=0.51; 95%CI:0.31–0.83; p=0.007). Thus, only the middle part of the distribution for lower lutein+zeaxanthin status was linked to reduced odds of this type of dyslipidemia, whereas the remaining part of the distribution (Q₃ and Q₄) showed a comparable odds of this outcome to Q₁. Similarly, a lower odds of EDS was found when comparing Q₂ to Q₁ of the low α-carotene gene score, but not others. In contrast, a gene score reflecting low lycopene level was associated with higher risk of central obesity only when comparing Q₂ to Q₁ (OR=2.81; 95%CI: 1.09–7.26; p=0.033) with the association weakening with each higher quartile comparison. Importantly, the lutein+zeaxanthin LSCRS was inversely related to the odds of having NCEP ATP III MetS though only for Q₂ and Q₃ vs. Q₁, without a signficant linear trend observed.

Associations of LTCRS with metabolic outcomes and EDS was assessed in Table 4, through a series of multiple logistic regression, using quartiles of the risk score as the main predictor and testing for linear trend in the association. Among key findings, an inverse and linear association between LTCRS and HDL-C dyslipidemia indicated that a low carotenoid status gene score was potentially protective against this outcome (Q₄ vs. Q₁: 0.67; 95%CI:0.45–0.99; p=0.046, P-trend=0.046). Similarly, a non-linear association was found for elevated CRP (Q₂ vs. Q₁: OR=0.63; 95%CI:0.43–0.91, p=0.015).

Table 4.

Low total carotenoid risk scores (LTCRS, quartiles) and their associations with selected binary metabolic outcomes (central obesity, NCEP ATP III MetS and its components, elevated HOMA-IR, elevated CRP and hyperuriciemia) and elevated depressive symptoms (EDS) among African-American adults: multiple logistic regression analysis¹

	Q1	Q2			Q3			Q4
		OR	95%CI	p-value	OR	95%CI	p-value	OR	95%CI	p-value
Obesity (P-trend=0.50)	1	0.86	(0.58;1.26)	0.44	1.11	(0.76;1.63)	0.59	0.80	(0.54;1.17)	0.25
Central obesity(P-trend=0.56)	1	0.85	(0.56;1.30)	0.45	1.01	(0.66;1.55)	0.96	0.83	(0.54;1.27)	0.38
NCEP ATP III MetS(P-trend=0.05)	1	0.87	(0.57;1.32)	0.51	0.71	(0.45;1.09)	0.12	0.68	(0.44;1.05)	0.08
Elevated blood pressure(P-trend=0.34)	1	0.86	(0.59;1.27)	0.46	0.70	(0.48;1.04)	0.08	0.88	(0.59;1.30)	0.52
Dyslipidemia, HDL-C(P-trend=0.046)	1	0.80	(0.54;1.19)	0.28	0.76	(0.51;1.12)	0.17	0.67	(0.45;0.99)	0.046
Dyslipidemia, TG(P-trend=0.18)	1	1.28	(0.81;2.02)	0.29	0.91	(0.56;1.46)	0.69	0.79	(0.48;1.29)	0.35
Hyperglycemia(P-trend=0.99)	1	0.93	(0.63;1.38)	0.71	1.14	(0.76;1.69)	0.50	0.93	(0.62;1.39)	0.73
Elevated HOMA-IR(P-trend=0.73)	1	0.90	(0.61;1.32)	0.59	1.25	(0.86;1.81)	0.24	0.82	(0.56;1.22)	0.34
Elevated CRP(P-trend=0.41)	1	0.63	(0.43;0.91)	0.015	0.75	(0.52;1.10)	0.14	0.79	(0.54;1.16)	0.23
Hyperuricemia(P-trend=0.48)	1	1.17	(0.77;1.78)	0.45	0.92	(0.60;1.42)	0.71	0.92	(0.60;1.43)	0.72
Elevated depressive symptoms (EDS, CES-D≥16) (P-trend=0.45)	1	1.10	(0.71;1.74)	0.65	1.25	(0.80;1.97)	0.33	1.16	(0.74;1.82)	0.52

Abbreviations: ABCG5=ATP-binding cassette, subfamily G, member 5; ApoA=Apolipoprotein A; ApoB=Apolipoprotein B; ApoE=Apolipoprotein E; ATP III=Adult Treatment Panel III; BCMO1=beta-carotene mono-oxygenase 1 enzyme; BCDO2=Beta-carotene di-oxygenase 2 enzyme; BMI=Body Mass Index; CD36=thrombospondin receptor; CES-D=Center for Epidemiologic Studies-Depression scale; CRP=C-reactive protein; DBP=Diastolic Blood Pressure; EDS=Elevated Depressive Symptoms; FABP2=Fatty Acid Binding Protein 2; GWAS=Genome wide association studies; HANDLS=Healthy Aging in Neighborhood of Diversity Across the Lifespan; HDL-C=High Density Lipoprotein-Cholesterol; LIPC=Hepatic Lipase; HOMA-IR= Homeostatic model assessment, Insulin Resistance; HWE=Hardy-Weinberg equilibrium; LD=Linkage disequilibrium; LDL-C=Low-density Lipoprotein-Cholesterol; LPL=Lipoprotein Lipase; LSCRS=Low Specific Carotenoid Risk Score; LTCRS=Low Total Carotenoid Risk Score; MDS=Multi-Dimensional Scaling; MetS=Metabolic Syndrome; NCEP=National Cholesterol Education Program; NHANES I=National Health and Nutrition Examination Surveys I; Q1=Lowest quartile; Q4=Uppermost quartile; SBP=Systolic Blood Pressure; SCARB1=Scavenger receptor class B member 1; SNP=Single Nucleotide Polymorphism; SNPlcar=Single Nucleotide Polymorphism for lower carotenoid status; TG=Triglycerides; WC=Waist circumference.

¹LTCRS was entered in the multiple logistic regression model (as quartiles, with first quartile being the referent category) as main predictors, to assess their net association with each of the metabolic outcomes and with elevated depressive symptoms. Covariate entered as potential confounder were: sex, age, poverty income ratio (<125% vs. ≥125%), education (<HS, HS, >HS), marital status (current, former or never, missing), smoking status (current, former or never, missing), drug use (current, past or never, missing) and 10 principal components to adjust for population structure.

Discussion

In this study, we examined associations of gene polymorphisms related to low carotenoid status with various metabolic outcomes and to EDS in an urban socioeconomically diverse sample of African-American adults. None of the key findings for single SNP analyses survived correction for multiple testing. However, an inverse association was found between LTCRS and HDL-C dyslipidemia. The β-cryptoxanthin LSCRS was associated with lower odds of HDL-C dyslipidemia, but higher odds of elevated depressive symptoms.

Previous studies examining SNP used in our SNPlcar focused on on dyslipidemia, type 2 diabetes, obesity and MetS. Particularly, SNPlcar_{1(ABCG5, rs6720173:C/G, lutein+zeaxanthin)}^{(21; 22)} was unrelated HDL-C dyslipidemia or other lipids based on a study of Puerto Rican adults, while other associations were for various other studied SNP on that gene locus.⁽⁴⁵⁾ ABCG5’s main function is to translocate various hydrophobic substrates including carotenoids and cholesterol across extra- and intracellular membranes.⁽²¹⁾

Moreover, only few studies directly examined SNPlcar_{2(Apo B-516, β-carotene)} ^{(22; 23)} in relation to lipid profile and other metabolic disturbances. In one study, while another ApoB SNP (rs676210) was associated with lowering of TG, SNPlcar₂ (rs934197:T/C) was not⁽⁴⁶⁾, a finding replicated by at least one other study. ⁽⁴⁷⁾ However, two recent studies detected an association between the “T” allele dosage of that SNP and higher postprandial TG level,⁽⁴⁸⁾ and increased insulin resistance.⁽⁴⁹⁾ In our study, prior to correction for multiple testing (Table 2 and OMS3), the ApoB-516 “C” allele dosage (SNPlcar_2(ApoB)) yielded an inverse association with MetS (OR=0.72; 95%CI:0.52–1.00; p=0.048) and elevated CRP (OR=0.70;95%CI:0.51–0.95, p=0.022) that was consistent with previous studies. However, this finding was against the hypothesized direction in which genetic polymorphisms linked to lower carotenoid status would be related to higher odds of metabolic outcomes and EDS. Apo-B is essential for chylomicra and/or very low-density lipoproteins (VLDL) assembly and secretion in the small intestine and the liver. Apo-B is also the main apolipoprotein of LDL-C, a major carrier of carotenoids and TG-rich lipoproteins.⁽⁵⁰⁾

ApoA-IV’s main function is lipid absorption and modifying lipoprotein size.⁽⁵¹⁾ Although no associations were detected in our study with SNPlcar₃ (Apo-A-IV, rs675:A/T), previously linked to lower serum lycopene^{(22; 23)}, other studies showed that this SNP was associated with fenofibrate’s ability to lower TG levels among non-MetS patients.⁽⁵²⁾

Moreover, in that same study,⁽⁵²⁾ one of the ApoE SNP included in SNPlcar₄ (rs429358:C/T) was associated with increased LDL-C with fenofibrate treatment in the MetS group. ApoE2 has established atheroprotective properties based on previous studies. (e.g.⁽⁵³⁾) However, we did not detect significant associations between SNPlcar_4(ApoE) and any of the outcomes studied.

BCMO1 and β,β-carotene-9′,10′-oxygenase (BCDO2), are involved in symmetric and asymmetric carotenoid cleavage respectively, and convert β-carotene and apocarotenals to retinal, thus influencing circulatory carotenoid levels. ⁽⁵⁶⁾ For two of the most highly studied SNP in the BCMO1 gene (SNPlcar₅: rs6564851:G/T and SNPlcar₆: rs6564851:T/G), with SNPlcar_{5(BCMO1, lutein+zeaxanthin)} and SNPlcar_{6(BCMO1, β-cryptoxanthin)}, there was no relation with metabolic outcomes or EDS, in accordance with a meta-analysis suggesting that the loss of BCMO1 function is unrelated with a higher type 2 diabetes risk.⁽⁵⁴⁾ Moreover, SNPlcar_{7(BCMO1, β-carotene)} and SNPlcar_{13(BCMO1, β-carotene)},⁽²⁸⁾ had no association with lower HDL-C level in a recent French Canadian Study.⁽⁵⁵⁾ No other SNPlcar on the BCMO1 gene locus were previously studied in relation to metabolic disturbance or depressive symptoms. Though none of the associations remained significant after correction for multiple testing, among notable associations prior to that correction, SNPlcar_{14(BCMO1, β-cryptoxanthin)}⁽²⁸⁾, was associated in our study with a higher odds of EDS (OR=2.05; 95%CI:1.27–3.31; p=0.003); (Table 2). Other associations were detected that were either in the expected direction (SNPlcar_{12(BCMO1, α-carotene)} and central obesity; SNPlcar_{14(BCMO1, β-cryptoxanthin)} and SNPlcar_{16(BCMO1, β-carotene)} with hypertension) or against the hypothesized direction (SNPlcar_{17(BCMO1, β-carotene)} and obesity; SNPlcar_{10(BCMO1, β-cryptoxanthin)} and hypertension). Thus, further larger studies are needed to reconcile those inconsistent findings within that gene locus.

SNPlcar_{19(CD36, lutein+zeaxanthin)} (thrombospondin receptor gene (rs13230419)) has been shown to increase the odds of MetS by 29–40% in the African-American population.⁽⁵⁷⁾ CD36 codes for a membrane protein that facilitates the uptake and utilization of fatty acids in key metabolic tissues. Our study found a similar putative effect, though only significant or marginally significant association prior to correction for multiple testing with MetS (OR=1.41, 95%CI:0.99,2.00, p=0.06) and dyslipidemia-TG (OR=0.66, 95% CI:0.46–0.94, p=0.021). In contrast, our study did not find any significant associations with SNPlcar₁₈. SNPlcar_{18(CD36, low lutein+zeaxnathin with more “A” alleles)}^{(22; 58)} was related to MetS in one previous case-control study of Egyptian adults with the “G” allele being more prevalent in cases (n=100) than in controls (n=100).⁽⁵⁹⁾ A similar finding was observed in another study of 317 African-American adults, whereby the “A” allele of CD36 (rs1761667:A/G) was associated with greater perceived creaminess regardless of fat content of salad dressings (p<0.01) and higher mean acceptance of added fats and oils (p=0.02) without a significant association with the obesity phenotype.⁽⁶⁰⁾

Hepatic lipase (LIPC) hydrolyzes triglycerides and phospholipids from high-, intermediate-, and low-density lipoproteins, transforming them into smaller and denser particles, and promoting the cellular uptake of HDL cholesterol.⁽⁶¹⁾ For the two LIPC gene SNPlcar (both rs1800588:T/C), SNPlcar₂₀ (TT vs. others) was previously shown to be associated with low α-carotene while SNPlcar₂₁ (C allele) was linked to lower β-carotene level.^{(22; 61)} A study conducted among a large cohort of Chinese adults (n=4,194) showed that the “T” allele was linked to higher HDL-C level than the “C” allele (p<0.0001). ⁽⁶²⁾ The same pattern was found in a large cohort study of Caucasian adults (n=4,662) with the “C” allele being associated with dyslipidemia-HDL-C (p<0.0001).⁽⁶³⁾ Our study failed to detect an association between LIPC SNPlcar and various outcomes of interest.

Fatty acid binding protein 2 (FABP2) related SNPlcar₂₂ (rs1799883:A/G, GG vs. others) was previously associated with lower serum lycopene level^{(22; 29)}. In a study of 315 elderly subjects with MetS who were of European descent, the “G” allele was linked to lower TG and higher HDL-C (p<0.05), indicative of lower risk for dyslipidemia of both types.⁽⁶⁴⁾ There were no notable associations between this polymorphism and any of our study outcomes of interest. FABP2 is an intracellular protein expressed only in the intestine, involved in the absorption and intracellular transport of dietary long chain fatty acids and carotenoids to their specific metabolic targets. ⁽⁶¹⁾

For lipoprotein lipase (LPL) related SNPlcar₂₃ (rs328:G/C; GG vs. CC, mainly low α-carotene ^{(22; 30)}), two previous studies conducted among Caucasian adults also indicated that the “C” allele was consistently linked to HDL-C dyslipidemia,^{(63; 65)} with one of them observing an additional link to TG-dyslipidemia.⁽⁶⁵⁾ However, a recent meta-analysis showed only a modest relationship between rs328:G/C and both types of dyslipidemia.⁽⁶⁶⁾ Prior to correction for multiple testing, our study was indicative of a consistent relationship in which the “G” allele was associated with a lower odds of elevated HOMA-IR (OR=0.66; 95%CI:0.44–0.98; p=0.037), (OSM 5). However, this SNPlcar was not found to be associated with any type of dyslipidemia in our study. LPL catalyses the hydrolysis of the TG component of circulating chylomicrons and very low density lipoproteins, in tissues other than liver, and indirectly affects the concentration of carotenoids. ⁽³⁰⁾

Scavenger receptor class B member 1 (SCARB1) SNPlcar₂₄ (SR-BI exon 1, rs61932577:A/G; GG vs. others), previously linked to a lower level of β-cryptoxanthin^{(22; 23)} was also studied in relation to lipid profiles among adults. The SRBI was shown to have a role in Apo-B-containing lipoproteins’ metabolism in animals and humans. In fact, SRBI constitutes a backup pathway to the usual LDL receptor-mediated pathways for the catabolism of these lipoproteins. This is particularly relevant in adults with high Apo-B containing lipoproteins, commonly occurring in patients with familial hypercholesterolemia.⁽⁶⁷⁾ Prior to correction for multiple testing, our study found that SNPlcar_24(SCARB1) (i.e. higher “G” allele dosage) was linked to a higher odds of obesity, but no association was found with HDL-C or TG dyslipidemia. The associations of SCARB1 with HDL-C and TG were previously investigated, with a higher dosage of “A” allele being related to higher HDL-C and lower LDL-C values in men, but not in women.⁽⁶⁸⁾ This finding was replicated in a large study of US Caucasians (Framingham study, 2463 nondiabetic and 187 diabetic), in which diabetic subjects with the less common allele (allele A) have lower lipid concentrations, particularly LDL-C.⁽⁶⁹⁾ Those two studies had a consistent pattern of association found in our present study, though with different outcomes. However, two other studies found no associations of this SNPlcar with various lipid parameters.^{(53; 70)} Inconsistent with the pattern of findings from our study and those of others, a study of 77 subjects who were heterozygous for familial hypercholesterolemia found that the “A” allele dosage of this SNP was associated with higher TG.⁽⁶⁷⁾

This study is to our knowledge, the first to systematically examine genetic polymorphisms previously shown to be associated with lower serum carotenoid levels in relation to metabolic disturbance and depressive symptoms in an urban population of African-American adults and to construct gene scores for that purpose. Despite its strengths, some limitations include a statistical power-limiting small sample size. Moreover, most GWAS yielding our SNP_lcar short list came from studies of subjects of European ancestry. Finally, serum carotenoid concentration data was lacking which prevented direct assessment of SNP_lcar associations with respective carotenoids and comparisons with previous studies of European ancestry subjects. Additionally, such data availability would have allowed using gene score weights depending on effect sizes of each SNP_lcar on various carotenoids. Finally, in few gene loci included in of gene score computations, a SNP was related to multiple carotenoids, specifically BCMO1. However, to by-pass this issue, the gene scores were made mutually exclusive by including only the most significant carotenoid for each of the carotenoid-specific gene score.

In conclusion, gene polymorphisms linked to low serum carotenoid status had mixed effects on metabolic disturbance and depression. Specifically, our findings do not support that low carotenoid status gene polymorphisms will necessarily lead to a poorer metabolic and depressive symptoms outcome. In fact, in most cases, the opposite trend was found, with the possible exception of the β-cryptoxanthin risk score and EDS. Therefore, there is a major discrepancy between what was found in studies linking serum carotenoids to metabolic disturbance and depressive symptoms and this study which used gene polymorphisms linked to low carotenoid status as the main exposure. It is possible that different carotenoids may interact either synergistically or antagonistically with each others to affect those outcomes. Thus, similar studies on larger African-American samples are needed to test gene-gene (epistasis) interactions between those carotenoid-related gene polymorphisms.

Supplementary Material

OSM

Supplemental Figure 1. LTCRS distribution among HANDLS African-American participants

Supplemental Figure 2. Dietary intakes of carotenoids based on average of two 24-hr recalls (wave 1, μg/1,000kcal/d) among HANDLS African-American participants with complete data on LTCRS: dot plot stratified by PIR and sex

Supplemental Figure 3. Dietary intakes of carotenoids based on average of two 24-hr recalls (wave 1, μg/1,000kcal/d) among HANDLS African-American participants with complete data on LTCRS: scatterplot with LTCRS

ABBREVIATIONS

ABCG5

ATP-binding cassette, subfamily G, member 5

ApoA

Apolipoprotein A

ApoB

Apolipoprotein B

ApoE

Apolipoprotein E

ATP III

Adult Treatment Panel III

BCMO1

beta-carotene mono-oxygenase 1 enzyme

BCDO2

Beta-carotene di-oxygenase 2 enzyme

BMI

Body Mass Index

CD36

thrombospondin receptor

CES-D

Center for Epidemiologic Studies-Depression scale

CRP

C-reactive protein

DBP

Diastolic Blood Pressure

EDS

Elevated Depressive Symptoms

FABP2

Fatty Acid Binding Protein 2

GWAS

Genome wide association studies

HANDLS

Healthy Aging in Neighborhood of Diversity Across the Lifespan

HDL-C

High Density Lipoprotein-Cholesterol

LIPC

Hepatic Lipase

HOMA-IR

Homeostatic model assessment, Insulin Resistance

HS

High School

HWE

Hardy-Weinberg equilibrium

LD

Linkage disequilibrium

LDL-C

Low-density Lipoprotein-Cholesterol

LPL

Lipoprotein Lipase

LSCRS

Low Specific Carotenoid Risk Score

LTCRS

Low Total Carotenoid Risk Score

MDS

Multi-Dimensional Scaling

MetS

Metabolic Syndrome

NCEP

National Cholesterol Education Program

NHANES I

National Health and Nutrition Examination Surveys I

Q1

Lowest quartile

Q4

Uppermost quartile

SBP

Systolic Blood Pressure

SCARB1

Scavenger receptor class B member 1

SNP

Single Nucleotide Polymorphism

SNPlcar

Single Nucleotide Polymorphism for lower carotenoid status

TG

Triglycerides

WC

Waist circumference