FAAH Pro129Thr Variant Is Associated with Increased Cholesterol Levels in Normal-Weight Metabolically Unhealthy Subjects

Abstract

Introduction:

The endocannabinoid system (ECS) plays an integral role in maintaining metabolic homeostasis, where an hyperactivation has been related with serum lipid alterations. The biological effects of ECS are limited by the activation of the endocannabinoid-degrading enzyme fatty acid amide hydrolase (FAAH) and by polyunsaturated fatty acid (PUFA) intake as precursors. The FAAH Pro129Thr variant has been associated with obesity in some populations. However, the association with metabolic phenotypes in the Mexican population has not been studied. This study aimed to analyze the association of the FAAH Pro129Thr variant with serum lipids and diet in Mexican adults with different metabolic phenotypes.

Methods:

This is a cross-sectional study with 306 subjects between 18 and 65 years of age. They were classified with normal weight (NW) or excess weight (EW) according to their body mass index (BMI). The EW group included individuals with overweight or obesity (BMI 25–39.9 kg/m²). The individuals were classified into two metabolic phenotypes, metabolically healthy and metabolically unhealthy (MUH), using the homeostatic model assessment of insulin resistance and the National Cholesterol Education Program-adenosine triphosphate III cutoff points for blood pressure, triglycerides, high-density lipoprotein cholesterol, and fasting glucose. Subjects with ≥2 of 5 altered parameters were classified as MUH. The FAAH Pro129Thr variant was determined by allelic discrimination with TaqMan^® probes.

Results:

The total cholesterol and very low-density lipoprotein cholesterol levels were associated with the FAAH Pro129Thr variant in NW-MUH subjects. Moreover, a lower PUFA intake was found in EW-MUH subjects with the FAAH variant.

Conclusions:

FAAH Pro129Thr variant has an important role in lipid metabolism, especially in NW-MUH subjects. By contrast, a low dietary intake of endocannabinoid PUFA precursors may partly counteract the development of the altered lipid profile associated with overweight/obesity.

Introduction

Excess weight (EW), defined as overweight or obesity, has become one of the biggest public health problems in Mexico as in the world,¹ and it has been associated with numerous adverse health outcomes, including cardiovascular disease, diabetes mellitus type 2 (DM2), and some types of cancer.²

However, EW is not a homogenous condition. For example, some people who present EW not always have metabolic abnormalities and could be considered to have metabolically healthy obesity. Therefore, they might be at lower risk of suffering cardiovascular events than those who present metabolic alterations.³ Nevertheless, there is no single universal definition to categorize metabolically healthy (MH) subjects. More than 30 different definitions of metabolic health have been described in clinical studies in which the absence of metabolic syndrome is frequently used as a classification criterion.^4,5

On the other hand, there are subjects with normal weight (NW) according to their body mass index (BMI <25 kg/m²), who may present metabolic alterations. Accordingly, these individuals present a metabolically unhealthy (MUH) phenotype characterized by multiple cardiometabolic disorders, such as chronic low-grade inflammation, insulin resistance, and high serum triglycerides (TG), blood pressure, fasting glucose, and decreased serum high-density lipoprotein cholesterol (HDL-c).⁶

In addition, MUH individuals tend to have altered lipid profiles with significantly higher serum levels of very low-density lipoprotein cholesterol (VLDL-c), intermediate-density lipoprotein cholesterol, and low-density lipoprotein cholesterol (LDL-c) compared to MH subjects.⁷

The multifactorial etiology of obesity is attributed to the interaction between multiple genetic variants and loci with the “obesogenic” environment. In addition, recent advances in genome-wide association studies have demonstrated that loci associated with obesity are located within genes involved in pathways affecting neurocircuits of appetite and satiety regulation, insulin secretion, energy, and lipid metabolism.⁸

The endocannabinoid system (ECS) is widely recognized as an essential regulator of feeding, energy expenditure, energy storage, appetite control, cardiovascular diseases, and inflammatory responses, being a target for obesity treatment.⁹

This system consists of two cannabinoid receptors (CB1 and CB2, most abundant in brain and peripheral organs, respectively), their endogenous ligands: anandamide (AEA) and 2-arachidonoylglycerol (2-AG), and the endocannabinoid-degrading enzymes (fatty acid amide hydrolase [FAAH] for AEA and monoacylglycerol lipase for 2-AG).¹⁰ Moreover, it has been shown that congeners of AEA and 2-AG, derived from several long chain fatty acids, including n-3 (e.g., α-linolenic, docosahexaenoic acid, and eicosapentaenoic acid) or n-6 (e.g., linoleic and arachidonic acid [AA]) polyunsaturated fatty acids (PUFAs), are also biosynthesized from cells, and that membrane phospholipids in which these fatty acids are esterified, are used as ultimate precursors for ECS and their congeners.⁹

The sustained CB1 receptor stimulation under different conditions induces glucose intolerance, stimulates metabolic endotoxemia, inflammation, and alters lipid and glucose metabolism in muscle, liver, and adipose tissue.^11,12 The ECS is primarily inactivated by cellular reuptake and intracellular hydrolysis.^13,14

FAAH is ubiquitous and is widely expressed in neuronal cells in the central nervous system.¹⁵ Due to the strong negative correlation between FAAH expression in adipose tissue and ECS levels in plasma, genetic variants have been studied. One of the most frequent is the 385A/C variant (rs324420) in human FAAH that replaces a conserved proline residue at amino acid position 129 to threonine. This replacement can decrease FAAH stability and activity, eventually resulting in a continuous stimulation of the CB1 receptor.¹⁶

Although genetic variation in FAAH was formerly associated with altered serum lipid profile,^17,18 others have failed to find such association.^19,20 In addition, to our knowledge, only one study on a Greek population cohort classified their participants concerning metabolic phenotypes and the FAAH Pro129Thr variant.²¹ Thus, this study aimed to analyze the association of the FAAH Pro129Thr variant with serum lipids and diet in Mexican adults with MH or MUH phenotype.

Materials and Methods

Subjects

This cross-sectional study of unrelated Mexican mestizos included subjects between 18 and 65 years of age. Height and weight were measured, and BMI was calculated for volunteers interested in participating. Adults with NW (BMI ≥18.5 to <25 kg/m²), overweight (BMI ≥25 to <30 kg/m²), or obesity (BMI ≥30 to <40 kg/m²) were included. A total of 443 adults were enrolled; however, only 306 subjects were included with a complete evaluation (anthropometry, diet, biochemistry, and genetic material) (Fig. 1). The study was conducted at the Institute of Translational Nutrigenetics and Nutrigenomics, University of Guadalajara, Jalisco, México.

FIG. 1.

Flowchart of study subjects.

Subjects were not included if they had any prescribed medication for chronic diseases, such as DM2, cardiovascular, liver, kidney, or pancreas, since such medication could alter blood pressure, serum glucose, or lipid profile. Women who were pregnant or breastfeeding were also not included.

Anthropometric and clinic measurements

Anthropometric parameters were measured after 8–10 hr of fasting. Measurements were performed with light clothes and without shoes. Tetrapolar electrical bioimpedance was used to determine body composition, including muscle and fat mass percentage (InBody 3.0; Biospace Co., Seoul, Korea). Height measurement was done using a stadiometer with a precision of 1 mm (Rochester Clinical Research, Inc., New York, NY, USA). BMI was calculated by dividing weight in kilograms by height in meters squared (kg/m²).²² Subjects were classified according to their BMI as NW (18.5 to <25 kg/m²) or EW if they presented overweight (≥25 to <30 kg/m²) or obesity (≥30 to <40 kg/m²).²³

Waist circumference was measurement in the narrowest diameter between the last rib and the iliac crest; besides, hip circumference was determined at the maximum posterior protuberance level of the gluteus using a Lufkin Executive^®Thinline 2 mm measuring tape (Lufkin Executive Thinline, W606PM, MD, USA). Systolic and diastolic blood pressure were measured with a LifeSource digital sphygmomanometer (LifeSource, Milpitas, CA, USA) after at least 15 min of rest. Subjects were instructed to sit with their back touching the chair and with the arm resting on a horizontal surface not crossing their legs. Two blood pressure measurements were made, and the average was registered.

Definition of MH and MUH phenotype

There is no consensus definition for the MUH phenotype. Nevertheless, in this study, for each BMI category, we considered the following criteria, established by Torres-Castillo et al.:²⁴ systolic blood pressure ≥130 mmHg and diastolic blood pressure ≥85 mmHg, TG ≥150 mg/dL, HDL-c <40 mg/dL in men and <50 mg/dL in women, fasting glucose ≥100 mg/dL, and the homeostasis model assessment of insulin resistance (HOMA-IR) >2.5.²⁵ If subjects had none or one of these altered cutoff points, they were considered MH phenotype, otherwise, they were classified as MUH phenotype (≥2 met criteria).

Therefore, individuals were grouped into one of four groups: NW-MH, NW-MUH, EW-MH, and EW-MUH, based on their BMI classification and metabolic condition.

Dietary intake

Subjects were provided with a 3-day food consumption record, which included a weekend day to estimate their habitual dietary intake. All the individuals were instructed to fill out the food records with the use of food scales and models from Nasco^® to enhance the accuracy of the portion sizes; then, the completed questionnaires were reviewed by a registered dietitian and analyzed using a specialized software (Nutrikcal VO^®, Mexico).

Biochemical analysis

Blood samples were taken after 8–10 hr of fasting and centrifuged to obtain the serum. Determinations of serum glucose, TG, total cholesterol (TC), and HDL-c were carried out by dry chemistry using a Vitros 350 Analyzer (Ortho-Clinical Diagnostics, Johnson & Johnson Services Inc., Rochester, NY, USA). The VLDL-c was calculated as TG/5 (mg/dL) and LDL-c was calculated using the Friedewald formula, as long as TG levels were <400 mg/dL²⁶ (LDL-c = TC – [HDL-c + VLDL-c]).

The cutoff points used to define alterations in the lipid profile were those established by the National Cholesterol Education Program III guidelines:²⁷ hypercholesterolemia ≥200 mg/dL, elevated LDL-c ≥100 mg/dL, and elevated VLDL-c ≥30 mg/dL.

Insulin levels were determined using an ELISA assay (Monobind Inc., Lake Forest, CA, USA) according to instructions from the supplier. Insulin resistance was estimated according to the HOMA-IR²⁵ and calculated as follows: [Fasting insulin (μU/mL) × fasting glucose (mg/dL)/405].

DNA extraction and genotyping

The genomic DNA was extracted from peripheral blood using the High Pure PCR Template Preparation kit (Roche Diagnostics, Mannheim, Germany) and then diluted to 20 ng/μL. FAAH rs324420 variant was determined by allelic discrimination using TaqMan^® probes (assay number C___1897306_10; Drug Metabolism Assay, Applied Biosystems, Foster City, CA, USA). The experiments were carried out in a Light-Cycler^® 96 Real-Time PCR System (Roche Diagnostics) under the following conditions: 95°C for 10 min and 40 cycles of denaturation at 95°C for 15 sec and annealing/extension at 60°C for 1 min. FAAH rs324420 genotyping was verified using positive controls of DNA samples corresponding to the three possible genotypes in each 96-well plate. Twenty percent of samples were analyzed in duplicate.

Statistical analysis

The statistical power was evaluated according to the calculation of the sample size, performed with an estimated error margin of 5% with a confidence degree of 95% and an expected prevalence of MH obese subjects of 19% reported in a previous study.²⁴ The Kolmogorov–Smirnov test was used to determine normal distribution of quantitative variables. All variables were log-transformed to better approximate a normal distribution. To analyze the differences between subjects NW or EW and MH or MUH phenotype, plus the FAAH genotype, the two-way analysis of covariance (ANCOVA) adjusted by sex, age, and energy intake was used. Variables were adjusted by sex, age, and kilocalories and are presented as the estimated mean and standard error of the mean. The Hardy–Weinberg equilibrium and the comparison of categorical variables were performed using the chi-squared test.

The association between metabolic variables and the FAAH variant was analyzed by calculating the odds ratio with a 95% confidence interval. Association analyzes were performed using the chi-squared test and logistic regression adjusted by age and sex. In addition, multiple linear regression analyses were performed, considering anthropometric, biochemical, and dietary as dependent variables and genotype, sex, age, and phenotypes as independent variables; collinearity was avoided by not including variables that were interrelated in the same model.

All statistical analyses were performed using SPSS v28.0 software (IBM Corp., Armonk, NY, USA), and a P value <0.05 was considered statistically significant.

Ethical considerations

Subjects were informed about the research procedures. All the participants provided signed written informed consent before enrollment to the study. Besides, this study was approved by the Research Ethical Committee of the University of Guadalajara (Register number: CI/019/2010), and it was carried out according to the Declaration of Helsinki principles.²⁸

Results

Characteristics of the study population according to FAAH Pro129Thr variant

The genotype distribution of the FAAH Pro129Thr variant across the entire sample (n = 306) was Pro129Pro 45.8% (n = 140), Pro129Thr 43.4% (n = 133), and Thr129Thr 10.8% (n = 33). The Thr risk allele frequency was 32.5%. The genotype frequencies were in Hardy–Weinberg equilibrium (P = 0.86). Moreover, no significant difference was found between the genotypes, the allelic frequency, or the dominant genetic model by BMI and metabolic phenotype (Table 1).

Table 1.

FAAH Genotypes and Allelic Frequencies in Subjects by Metabolic Phenotypes

FAAH Pro129Thr variant	NW-MH (n = 83), % (n)	NW-MUH (n = 37), % (n)	EW-MH (n = 61), % (n)	EW-MUH (n = 125), % (n)	P	OR (95% CI)	Total (n = 306), % (n)
Genotype
Pro/Pro	43.4 (36)	51.4 (19)	45.9 (28)	45.6 (57)		1	45.8 (140)
Pro/Thr	41.0 (34)	35.1 (13)	49.2 (30)	44.8 (56)	0.71	1.04 (0.85–1.26)	43.5 (133)
Thr/Thr	15.7 (13)	13.5 (5)	4.9 (3)	9.6 (12)	0.81	0.81 (0.60–1.09)	10.8 (33)
Allele
Pro	63.8 (106)	68.9 (51)	70.5 (86)	68 (170)	0.61	1	67.5 (213)
Thr	36.2 (60)	31.1 (23)	29.5 (36)	32 (80)	0.61	0.91 (0.65–1.28)	32.5 (199)
p-HWE	0.30	0.27	0.15	0.74			0.86
Dominant genetic model
Pro/Pro	43.4 (36)	51.4 (19)	45.9 (28)	45.6 (57)	0.88	1	45.8 (140)
Pro/Thr + Thr/Thr	56.6 (47)	48.6 (18)	54.1 (33)	54.4 (68)	0.88	0.98 (0.82–1.18)	54.2 (166)

The P value corresponds to chi square test.

CI, confidence interval; EW-MH, excess weight-metabolically healthy phenotype; EW-MUH, excess weight-metabolically unhealthy phenotype; FAAH, fatty acid amide hydrolase; NW-MH, normal weight-metabolically healthy phenotype; NW-MUH, normal weight-metabolically unhealthy phenotype; OR, odds ratio; p-HWE, P value of Hardy–Weinberg equilibrium.

Characteristics of subjects by metabolic phenotype, and FAAH genotypes

Among all the subjects, the mean age was 37.3 ± 11.3 years, and 74.5% (n = 228) were women. Based on BMI, 120 subjects (39.2%) were classified as NW and 186 (60.8%) as EW. With respect to EW subjects, 67.2% (n = 125) had MUH phenotype, while this phenotype was present in 30.8% (n = 37) of the NW subjects. In addition, sociodemographic, anthropometric, biochemical, and dietetic characteristics are presented in Table 2.

Table 2.

General Characteristics of Subjects Analyzed by Normal Weight or Excess Weight, Metabolic Phenotype, and FAAH Genotype

	NW-MH (n = 83, 27%)		EW-MH (n = 61, 20%)		NW-MUH (n = 37, 12%)		EW-MUH (n = 125, 41%)		P
	Genotype Pro/Pro (n = 36)	Genotype Pro/Thr + Thr/Thr (n = 47)	Genotype Pro/Pro (n = 28)	Genotype Pro/Thr + Thr/Thr (n = 33)	Genotype Pro/Pro (n = 19)	Genotype Pro/Thr + Thr/Thr (n = 18)	Genotype Pro/Pro (n = 57)	Genotype Pro/Thr + Thr/Thr (n = 68)	Phenotype	Genotype	Interaction
Sociodemographic variables
Women (%)	80.55	78.72	67.85	78.79	57.90	66.67	80.7	70.59	—	—	—
Age (years)	33.83 ± 1.82	31.11 ± 1.59	39 ± 2.07	38.3 ± 1.90	36.47 ± 2.51	41.5 ± 2.58	40.4 ± 1.45	38.63 ± 1.32	—	—	—
Anthropometric variables
Weight (kg)	57.55 ± 2.04	58.55 ± 1.82	63.14 ± 2.80	64.17 ± 2.88	76.53 ± 2.30	73.89 ± 2.12	86.28 ± 1.63	86.94 ± 1.48	<0.001	0.851	0.772
BMI (kg/m²)	21.29 ± 0.74	22.02 ± 0.66	23.48 ± 1.01	23.22 ± 1.04	29.28 ± 0.83	27.72 ± 0.76	32.66 ± 0.59	32.90 ± 0.53	<0.001	0.783	0.338
Muscle mass (kg)	10.94 ± 0.28	10.92 ± 0.27	11.07 ± 0.39	11.47 ± 0.40	12.20 ± 0.32	12.40 ± 0.31	13.63 ± 0.22	13.22 ± 0.20	<0.001	0.919	0.588
Fat mass M (%)	17.53 ± 2.28	19.38 ± 1.96	19.92 ± 2.12	20.38 ± 2.56	24.79 ± 2.02	23.82 ± 2.27	26.37 ± 1.81	28.22 ± 1.34	<0.001	0.523	0.804
Fat mass W (%)	25.96 ± 1.01	27.05 ± 0.91	30.74 ± 1.62	30.16 ± 1.56	38.94 ± 1.24	36.59 ± 1.06	41.08 ± 0.80	41.71 ± 0.78	<0.001	0.765	0.531
WC M (cm)	82.60 ± 3.84	83.72 ± 3.29	84.76 ± 3.57	88.09 ± 4.31	94.11 ± 3.40	98.31 ± 3.81	102.13 ± 3.04	103.62 ± 2.25	<0.001	0.272	0.967
WC W (cm)	71.80 ± 1.92	74.14 ± 1.73	79.10 ± 3.10	78.44 ± 3.10	91.75 ± 2.36	87 ± 2.02	101.66 ± 1.53	100.28 ± 1.49	<0.001	0.582	0.312
Biochemical variables
TG (mg/dL)	95.84 ± 10.97	93.87 ± 9.78	98.98 ± 12.38	108.36 ± 11.39	131.60 ± 15.07	207.75 ± 15.49	185.01 ± 8.74	191.16 ± 7.95	<0.001	0.005	0.047
HDL-c M (mg/dL)	51.49 ± 3.13	45.01 ± 2.69	43.94 ± 2.77	42.58 ± 3.11	36.53 ± 2.91	36.51 ± 3.52	33.45 ± 2.48	35.19 ± 1.84	<0.001	0.413	0.519
HDL-c W (mg/dL)	56.63 ± 1.93	54.94 ± 1.74	50.56 ± 2.37	49.13 ± 2.03	40.56 ± 3.11	42.40 ± 2.98	40.13 ± 1.54	38.34 ± 1.49	<0.001	0.494	0.963
TC (mg/dL)	178.59 ± 5.77	178.15 ± 5.14	189.18 ± 6.51	175.95 ± 5.99	178.09 ± 7.93^a	205.54 ± 8.38^a	188.94 ± 4.60	190.96 ± 4.21	0.146	0.455	0.045
LDL-c (mg/dL)	104.55 ± 5.17	108.56 ± 4.63	120.83 ± 5.75^a	106.96 ± 5.29^a	111.15 ± 7.18	115.60 ± 7.42	112.79 ± 4.09	115.88 ± 3.69	0.555	0.689	0.162
VLDL-c (mg/dL)	19.09 ± 2.05	18.63 ± 1.83	19.36 ± 2.41	21.66 ± 2.13	26.29 ± 2.82^a	41.66 ± 2.90^a	35.71 ± 1.67	37.66 ± 1.51	<0.001	0.003	0.042
Dietary intake
Energy (Kcal)	1904.4 ± 157.1	1745.5 ± 138.9	1850.2 ± 181.7	2237.2 ± 155.9	2135.8 ± 155.9	2030.9 ± 143.3	2205.9 ± 102.4	2188.8 ± 95.1	0.033	0.835	0.637
Carbohydrates (%)	49.55 ± 1.98	50.45 ± 1.77	47.55 ± 2.29	52.99 ± 3.03	47.25 ± 1.96	48.87 ± 1.80	45.47 ± 1.29	47.88 ± 1.2	0.167	0.105	0.759
Protein (%)	15.99 ± 0.87	14.19 ± 0.78	15.78 ± 1.01	15.98 ± 1.34	18.73 ± 0.87	15.72 ± 0.79	17.18 ± 0.57	16.88 ± 0.53	0.029	0.056	0.197
FAT (%)	35.45 ± 1.89	35.35 ± 1.70	36.65 ± 2.19	31.03 ± 2.90	34 ± 1.88	35.39 ± 1.72	37.36 ± 1.24	35.24 ± 1.15	0.632	0.235	0.466
SFA (%)	9.20 ± 0.73	9.73 ± 0.65	10.87 ± 0.85	10.14 ± 1.13	10.84 ± 0.88	10.06 ± 0.75	9.58 ± 0.55	9.16 ± 0.50	0.190	0.479	0.963
MUFA (%)	10.65 ± 0.95	11.88 ± 0.85	12.17 ± 1.10	10.71 ± 1.47	11.78 ± 1.15	14.12 ± 0.97	12.48 ± 0.71^a	10.45 ± 0.66^a	0.264	0.911	0.072
PUFA (%)	4.48 ± 0.49	5.61 ± 0.44	5.61 ± 0.57	4.66 ± 0.76	4.48 ± 0.60	5.75 ± 0.51	5.73 ± 0.37^a	4.76 ± 0.34^a	0.971	0.823	0.013
Omega-6	14.99 ± 2.36	12.18 ± 2.10	15.01 ± 2.95	13.84 ± 4.10	15.17 ± 2.67	13.08 ± 4.10	15.98 ± 2.86	14.80 ± 2.35	0.970	0.374	0.965
Omega-3	1.42 ± 0.21	1.25 ± 0.19	1.58 ± 0.26	1.35 ± 0.21	1.55 ± 0.24	1.61 ± 0.37	1.59 ± 0.26	1.32 ± 0.21	0.955	0.202	0.731
Omega 6:3 ratio	10.59 ± 1.46	10.31 ± 1.29	11.01 ± 1.82	10.74 ± 1.41	9.97 ± 1.64	8.75 ± 2.52	10.53 ± 1.76	10.33 ± 1.45	0.728	0.429	0.689

All data are presented as mean ± SEM and were calculated according to two-way ANCOVA adjusted by sex, age, and BMI; in the case of dietary variables, data were adjusted by total energy as well. All P values were calculated with log-transformed variables for the analysis. P values were calculated between phenotypes (MH vs. MUH), genotypes (dominant genetic model Pro/Pro vs. Pro/Thr + Thr/Thr), and their interactions. Bold numbers represent statistical significance (P < 0.05).

Indicates the differences between different genotypes (Pro/Pro vs. Pro/Thr + Thr/Thr) within the same phenotype (NW-MH, EW-MH, NW-MUH, and EW-MUH) through a t Student test.

ANCOVA, analysis of covariance; BMI, body mass index; FAT, fatty acids totals; HDL-c, high-density lipoprotein cholesterol; LDL-c, low-density lipoprotein cholesterol; M, men; MH, metabolically healthy; MUFA, monounsaturated fatty acids; MUH, metabolically unhealthy; PUFA, polyunsaturated fatty acids; SEM, standard error of the mean; SFA, saturated fatty acids; TC, total cholesterol; TG, triglycerides; VLDL-c, very low-density lipoprotein cholesterol; W, women; WC, waist circumference.

The study variables were analyzed by genotype and metabolic phenotype in subjects with NW and EW. According to MH/MUH phenotype, we observed significant differences in all the anthropometric variables, VLDL-c, and energy. On the other hand, in VLDL-c, the differences by genotypes (Pro/Pro vs. Pro/Thr + Thr/Thr) were statistically significant.

Also, there were statistical differences in TC, VLDL-c, and PUFAs related to genotype and phenotype interaction (Table 2).

Association of FAAH Pro129Thr variant with lipids

Because significant statistical differences were observed in P values for genotype and phenotype interaction in TC (P = 0.045), VLDL-c (P = 0.042), and PUFAs (P = 0.013) (Table 2), subsequent analyses were performed to evaluate possible associations. Multivariable linear regression analyses were conducted to analyze the associations of TC, VLDL-c, and PUFAs with the FAAH Pro129Thr variant (dominant model Pro/Pro vs. Pro/Thr + Thr/Thr). In subjects with NW-MUH, models showed a positive and significant association between the presence of the Thr risk allele with TC and VLDL-c after adjusting by sex and age (Table 3).

Table 3.

Association of the FAAH Variant with Serum Lipids

Model	Dependent variable	R ²	β (95% CI)	P
NW-MUH
Model 1	TC (mg/dL)	0.12	30.84 (2.17–59.52)	0.036
Model 2	VLDL-c (mg/dL)	0.14	15.39 (2.19–28.58)	0.024

Multiple linear regression models between FAAH variant (dominant model Pro/Pro vs. Pro/Thr + Thr/Thr) as independent variable, and serum lipids.

In addition, the NW-MUH subjects with the risk allele had a higher risk for elevated VLDL-c serum levels (Table 4).

Table 4.

Association of the FAAH Pro129Thr Variant with Elevated Levels of Very Low-Density Lipoprotein Cholesterol

Subjects	OR	β (CI 95%)	P
NW-MUH	5.633	1.72 (1.37–23.16)	0.017

Logistic regression of FAAH variant using dominant model (Pro/Pro vs. Pro/Thr + Thr/Thr). Adjusted for sex and age.

Discussion

In this study, we evaluated the association between the FAAH Pro129Thr variant with serum lipids and diet in Mexican adults with MH or MUH phenotypes. To our knowledge, this study is the first to analyze the association between metabolic phenotypes with the FAAH Pro129Thr variant in Mexican population.

Our results indicate that higher TC and VLDL-c levels were associated with the risk Thr allele in NW-MUH subjects. Also, PUFA intake was lower in subjects with the risk allele in EW-MUH subjects. Results did not change after adjusting for intervening variables, including age and sex.

In a study of 1340 lean normoglycemic French adults, significantly higher levels of HDL-c were observed, in addition to a trend toward higher levels of TC, in subjects with the Thr/Thr genotype.²⁹ Furthermore, in another study of 121 lean Greek individuals, the control and MH obesity groups reported an association with higher HDL-c levels in the risk allele carrier subjects.²¹

Aberle et al. reported in a study of 451 adults of Caucasian origin that, after 6 weeks of a low-fat diet (40–50 g/daily), a significant reduction in TG and TC in carriers of the variant allele was observed.³⁰

According to evidence, people with the FAAH Pro129Thr variant produce a functionally deficient protein with about half the enzymatic activity compared to the wild type, and, since FAAH is the main AEA-degrading protein, this impairment is expected to lead to an increase in AEA (and related N-acylethanolamine) levels both peripherally and centrally.³¹

The relationship between ECS activity with lipid metabolism and energy consumption has been recognized. There is also a relationship between CB1 stimulation and the activation of genes related to lipogenesis and fatty acid synthesis.^32
–34

Thus, in mice, the activation of CB1 in the liver due to higher levels of AEA induces the expression of the lipogenic transcription factor sterol regulatory element binding protein (SREBP-1c) and its target enzymes, acetyl coenzyme-A carboxylase-1 (Acaca) and fatty acid synthase (Fas), and also increases de novo fatty acid synthesis.³³ Accordingly, in mice with diet-induced obesity, it was observed that the blockade of CB1 by AM6545, a neutral CB1-receptor antagonist that does not penetrate the brain, causes a decrease in stearoyl coenzyme-A desaturase 1 (Scd1), Fas, and Acaca, and an increase in carnitine palmitoyltransferase-1 (Cpt1).³⁴

Moreover, the ECS ultimately originates from AA through the action of N-acyl phosphatidylethanolamine-selective phospholipase D (NAPE-PLD) or diacylglycerol lipases (DAGL) α and β on arachidonate-containing biosynthetic precursors derived from membrane phospholipids. Dietary PUFAs from the n-6 and n-3 series determine the amount of AA esterified in phospholipids.³⁵ It has been described that AA supplementation increases AEA levels,³⁶ while dietary n-3 PUFAs reduce the concentrations of AA and ECS in rodent tissues and human plasma.^37
–39 This demonstrates that PUFA intake can determine the potential amounts of PUFA-derived lipid mediators produced in tissues.^35,40

Our results show that EW-MUH subjects with the Thr risk allele had a lower intake of PUFAs compared to subjects with the wild-type allele. This finding, although not directly related to the genetic variant, indirectly suggests that these subjects could have lower ECS, since, despite having a deficient FAAH enzyme, their ECS profile may not be altered in the same way because intake of ECS precursors (PUFA) is lower; however, further studies are needed to confirm these results and the relationship between PUFA intake and circulating ECS.

On the other hand, regarding the metabolic phenotypes, it can be observed that MUH subjects with the risk allele had significantly higher serum lipid levels. These results indicate that the variant influences metabolic health; therefore, it is essential to consider that the analyses of phenotype (not only BMI) and FAAH variant could be important to prevent chronic diseases and to give better nutritional and medical treatment if necessary.

The possible discrepancies observed in different studies regarding the association of the FAAH variant with biochemical or dietary variables could be related to the effect of other ECS mediators. Disruption of the FAAH enzyme alters the metabolism of the entire N-acylethanolamine family, resulting in increases of not only AEA but also oleoylethanolamide and palmitoylethanolamide levels, which have opposing metabolic effects.⁴¹ Also, other gene-gene or gene-environment interactions could influence the appearance of metabolic alterations.

Finally, in addition to the study of genetic variants, it is important to consider that in some obesity-related traits, there are physiological factors, such as age and sex, as well as environmental modulators, including physical activity, diet, ethnicity, and socioeconomic and educational status, which may increase the risk of developing an MUH phenotype.⁴²

The main strength of this study is the classification of subjects by metabolic phenotypes, considering multiple variables to ensure homogeneous groups for comparison. Furthermore, our results were adjusted for possible confounding variables. To our knowledge, this is the first study on Mexican population that associates metabolic phenotypes with FAAH genotypes. We expect that this study will open the possibility of conducting similar research in other ethnic groups.

One of the limitations of this study, as other similar studies, is the heterogeneous definitions for MUH that make it difficult to compare with other studies. Moreover, our results are restricted to Western Mexico population; thus, additional studies are required to compare our findings in different regions and genetic populations. In addition, although other studies have already demonstrated that the FAAH Pro129Thr variant is associated with higher circulating AEA levels,¹⁷ N-acylethanolamine and ECS profiles could not be measured in this study; however, it is important to consider these measurements in further studies as well as PUFAs to analyze the relationship as precursor molecules.

Therefore, a direct relationship between the metabolic traits found and potential enzymatic alteration-mediated effects on ECS and endocannabinoid-like molecules could not be established. Particularly, it would be interesting to see in EW-MUH subjects if the concomitant presence of the FAAH Pro129Thr variant and reduced PUFA intake result in no net change in AEA levels.

Finally, n-6 and n-3 consumption, and its ratio were measured; however, we did not find differences between groups, neither NW and EW nor MH and MUH. As a perspective, it would be important to analyze the relationship of ECS with PUFA intake in subjects with optimal consumption versus subjects with lower PUFA intake accompanied by a higher n-6/n-3 ratio, which has been reported as the most common pattern in this population.⁴³

Conclusions

Our study provides new evidence suggesting that the FAAH Pro129Thr variant has an important role in lipid metabolism, especially in NW-MUH subjects showing a positive and significant association between the presence of Thr risk allele with TC and VLDL-c. Further studies are needed to confirm the relationship between ECS and n-6 and n-3 PUFA intake as precursor molecules.

Footnotes

Acknowledgments

The authors thank the University of Guadalajara, and the subjects who participated in the study.

Authors' Contributions

E.S.R.: conceptualization (equal), formal analysis (lead), investigation (lead), methodology (equal), and original draft (lead). N.T.C.: data curation (equal), investigation (supporting), and review and editing (equal). B.V.: methodology (equal), resources (equal), original draft (supporting), and review and editing (equal). W.C.P.: data curation (equal), investigation (supporting), and review and editing (equal). O.D.L.C.: review and editing (equal). V.D.: supervision (supporting) and review and edition (equal). E.M.L.: conceptualization (equal), methodology (equal), resources (equal), original draft (supporting), and review and editing (equal).

Author Disclosure Statement

The authors declare that they have no competing interests and no personal financial interests. The authors declare that they are not currently employed, nor have they been employed (in the last 5 years), in an organization that stands to gain or lose financially from the publication of this article.

Funding Information

This research was funded, in part, by PROINPEP 2018 Universidad de Guadalajara, grant for E.S.R., B.V., and E.M.L. We also benefit from the support of the Program for Strengthening Research and Postgraduate Studies 2020 (Programa de Fortalecimiento de la investigación y el posgrado 2020; REC/0342/2020) grant to E.M.L. and B.V., and PRO-SNI 2018 (grant to B.V.).

References

Barquera

, Hernández-Barrera

, Trejo-Valdivia

, et al. Obesity in Mexico, prevalence and trends in adults. Ensanut 2018–19. Salud Publica Mex, 2020; 62(6):682–692.

Kelly

, Yang

, Chen

C-S

, et al. Global burden of obesity in 2005 and projections to 2030. Int J Obes, 2008; 32(9):1431–1437; doi: 10.1038/ijo.2008.102

Phillips

. Metabolically healthy obesity: Definitions, determinants and clinical implications. Rev Endocr Metab Disord, 2013; 14(3):219–227; doi: 10.1007/s11154-013-9252-x

Eckel

, Meidtner

, Kalle-Uhlmann

, et al. Metabolically healthy obesity and cardiovascular events: A systematic review and meta-analysis. Eur J Prev Cardiol, 2016; 23(9):956–966; doi: 10.1177/2047487315623884

Rey-López

, de Rezende

, Pastor-Valero

, et al. The prevalence of metabolically healthy obesity: A systematic review and critical evaluation of the definitions used. Obes Rev, 2014; 15(10):781–790; doi: 10.1111/obr.12198

Buscemi

, Chiarello

, Buscemi

, et al. Characterization of metabolically healthy obese people and metabolically unhealthy normal-weight people in a general population cohort of the ABCD study. J Diabetes Res, 2017; 2017:9294038; doi: 10.1155/2017/9294038

Telle-Hansen

, Christensen

, Formo

, et al. A comprehensive metabolic profiling of the metabolically healthy obesity phenotype. Lipids Health Dis, 2020; 19(1):90; doi: 10.1186/s12944-020-01273-z

Rohde

, Keller

, la Cour Poulsen

, et al. Genetics and epigenetics in obesity. Metabolism, 2019; 92:37–50; doi: 10.1016/j.metabol.2018.10.007

Freitas

, Isaac

, Malcher-Lopes

, et al. Polyunsaturated fatty acids and endocannabinoids in health and disease. Nutr Neurosci, 2018; 21(10):695–714; doi: 10.1080/1028415X.2017.1347373

10.

Di Marzo

. New approaches and challenges to targeting the endocannabinoid system. Nat Rev Drug Discov, 2018; 17(9):623–639; doi: 10.1038/nrd.2018.115

11.

Geurts

, Muccioli

, Delzenne

, et al. Chronic endocannabinoid system stimulation induces muscle macrophage and lipid accumulation in type 2 diabetic mice independently of metabolic endotoxaemia. PLoS One, 2013; 8(2):e55963; doi: 10.1371/journal.pone.0055963

12.

Di Marzo

, Silvestri

. Lifestyle and metabolic syndrome: Contribution of the endocannabinoidome. Nutrients, 2019; 11(8):1956; doi: 10.3390/nu11081956

13.

Cravatt

, Giang

, Mayfield

, et al. Molecular characterization of an enzyme that degrades neuromodulatory fatty-acid amides. Nature, 1996; 384(6604):83–87; doi: 10.1038/384083a0

14.

Simard

, Archambault

A-S

, Lavoie

J-PC

, et al. Biosynthesis and metabolism of endocannabinoids and their congeners from the monoacylglycerol and N-acyl-ethanolamine families. Biochem Pharmacol, 2022; 205:115261; doi: 10.1016/j.bcp.2022.115261

15.

Waleh

, Cravatt

, Apte-Deshpande

, et al. Transcriptional regulation of the mouse fatty acid amide hydrolase gene. Gene, 2002; 291(1):203–210; doi: 10.1016/S0378-1119(02)00598-X

16.

Yagin

, Aliasgari

, Aliasgharzadeh

, et al. The influence of the fatty acid amide hydrolase 385C>A single nucleotide polymorphisms on obesity susceptibility. Mol Biol Rep, 2019; 46(5):5049–5055; doi: 10.1007/s11033-019-04956-8

17.

Zeng

, Li

, Huang

. 385 C/A polymorphism of the fatty acid amide hydrolase gene is associated with metabolic syndrome in the Chinese Han population. Arch Med Sci, 2011; 7(3):423–427; doi: 10.5114/aoms.2011.23406

18.

Zhang

, Sonnenberg

, Baye

, et al. Obesity-related dyslipidemia associated with FAAH, independent of insulin response, in multigenerational families of Northern European descent. Pharmacogenomics, 2009; 10(12):1929–1939; doi: 10.2217/pgs.09.122

19.

Martins CJ de

, Genelhu

, Pimentel

MMG

, et al. Circulating endocannabinoids and the polymorphism 385C>A in fatty acid amide hydrolase (FAAH) gene may identify the obesity phenotype related to cardiometabolic risk: A Study Conducted in a Brazilian Population of Complex Interethnic Admixture. PLoS One, 2015; 10(11):e0142728; doi: 10.1371/journal.pone.0142728

20.

de Luis

, Sagrado

, Pacheco

, et al. Effects of C358A missense polymorphism of the endocannabinoid degrading enzyme fatty acid amide hydrolase on weight loss and cardiovascular risk factors 1 year after biliopancreatic diversion surgery. Surg Obes Relat Dis, 2010; 6(5):516–520; doi: 10.1016/j.soard.2010.01.005

21.

Papazoglou

, Panagopoulos

, Papanas

, et al. The fatty acid amide hydrolase (FAAH) Pro129Thr polymorphism is not associated with severe obesity in Greek subjects. Horm Metab Res, 2008; 40(12):907–910; doi: 10.1055/s-0028-1087169

22.

Anonymous. Obesity: Preventing and Managing the Global Epidemic. Report of a WHO Consultation. 2000;894:i–xii, 1–253. World Health Organization Technical Report Series, 2000; 894:i–xii, 1–253.

23.

Anonymous. Obesity and Overweight. n.d. Available from: https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight [Last accessed: May 18, 2022 ].

24.

Torres-Castillo

, Campos-Perez

, Gonzalez-Becerra

, et al. Waist circumference is an anthropometric parameter that identifies women with metabolically unhealthy phenotypes. Nutrients, 2018; 10(4):E447; doi: 10.3390/nu10040447

25.

Matthews

, Hosker

, Rudenski

, et al. Homeostasis model assessment: Insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia, 1985; 28(7):412–419; doi: 10.1007/BF00280883

26.

Friedewald

, Levy

, Fredrickson

. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem, 1972; 18(6):499–502.

27.

Expert Panel on

Detection

, Evaluation, and Treatment of High Blood Cholesterol in

Adults

. Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA, 2001; 285(19):2486–2497; doi: 10.1001/jama.285.19.2486

28.

World Medical Association. World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA, 2013; 310(20):2191–2194; doi: 10.1001/jama.2013.281053

29.

Durand

, Lecoeur

, Delplanque

, et al. Evaluating the association of FAAH common gene variation with childhood, adult severe obesity and type 2 diabetes in the French population. Obes Facts, 2008; 1(6):305–309; doi: 10.1159/000178157

30.

Aberle

, Fedderwitz

, Klages

, et al. Genetic variation in two proteins of the endocannabinoid system and their influence on body mass index and metabolism under low fat diet. Horm Metab Res, 2007; 39(5):395–397; doi: 10.1055/s-2007-977694

31.

Sipe

, Chiang

, Gerber

, et al. A missense mutation in human fatty acid amide hydrolase associated with problem drug use. Proc Natl Acad Sci U S A, 2002; 99(12):8394–8399; doi: 10.1073/pnas.082235799

32.

Lichtman

, Cravatt

. Food for thought: Endocannabinoid modulation of lipogenesis. J Clin Invest, 2005; 115(5):1130–1133; doi: 10.1172/JCI200525076

33.

Osei-Hyiaman

, DePetrillo

, Pacher

, et al. Endocannabinoid activation at hepatic CB1 receptors stimulates fatty acid synthesis and contributes to diet-induced obesity. J Clin Invest, 2005; 115(5):1298–1305; doi: 10.1172/JCI200523057

34.

Tam

, Vemuri

, Liu

, et al. Peripheral CB1 cannabinoid receptor blockade improves cardiometabolic risk in mouse models of obesity. J Clin Invest, 2010; 120(8):2953–2966; doi: 10.1172/JCI42551

35.

Watkins

, Hutchins

, Li

, et al. The endocannabinoid signaling system: A marriage of PUFA and musculoskeletal health. J Nutr Biochem, 2010; 21(12):1141–1152; doi: 10.1016/j.jnutbio.2010.04.011

36.

Berger

, Crozier

, Bisogno

, et al. Anandamide and diet: Inclusion of dietary arachidonate and docosahexaenoate leads to increased brain levels of the corresponding N-acylethanolamines in piglets. Proc Natl Acad Sci USA, 2001; 98:6402–6406; doi: 10.1073/pnas.101119098

37.

Gastón

, María Eugenia

, Das

, et al. Polyunsaturated fatty acids differentially modulate cell proliferation and endocannabinoid system in two human cancer lines. Arch Med Res, 2017; 48(1):46–54; doi: 10.1016/j.arcmed.2017.01.009

38.

Berge

, Piscitelli

, Hoem

, et al. Chronic treatment with krill powder reduces plasma triglyceride and anandamide levels in mildly obese men. Lipids Health Dis, 2013; 12:78; doi: 10.1186/1476-511X-12-78

39.

Batetta

, Griinari

, Carta

, et al. Endocannabinoids may mediate the ability of (n-3) fatty acids to reduce ectopic fat and inflammatory mediators in obese Zucker rats. J Nutr, 2009; 139(8):1495–1501; doi: 10.3945/jn.109.104844

40.

Castonguay-Paradis

, Lacroix

, Rochefort

, et al. Dietary fatty acid intake and gut microbiota determine circulating endocannabinoidome signaling beyond the effect of body fat. Sci Rep, 2020; 10(1):15975; doi: 10.1038/s41598-020-72861-3

41.

Sipe

, Scott

, Murray

, et al. Biomarkers of endocannabinoid system activation in severe obesity. PLoS One, 2010; 5(1):e8792; doi: 10.1371/journal.pone.0008792

42.

Torres-Castillo

, Martinez-Lopez

, Vizmanos-Lamotte

, et al. Healthy obese subjects differ in chronotype, sleep habits, and adipose tissue fatty acid composition from their non-healthy counterparts. Nutrients, 2020; 13(1):E119; doi: 10.3390/nu13010119

43.

Campos-Pérez

, González-Becerra

, Ramos-López

, et al. Same dietary but different physical activity pattern in normal-weight and overweight Mexican subjects. J Food Nutr Res, 2016; 4(11):729–735; doi: 10.1269/1/jfnr-4-11-5