The Associations of Erythrocyte Fatty Acids with Whole Blood Mineral Elements in Children

Background. Minerals play important biological roles in lipid metabolism. The primary aim of this study was to examine the relationships between erythrocyte fatty acids (FAs) levels with whole blood mineral elements concentrations among Chinese children. Methods. A cross-sectional study was conducted. A total of 435 children aged 4–7 years were recruited. Whole blood mineral elements were determined by atomic absorption spectrometry and erythrocyte FAs composition by gas chromatography-mass spectrometer. Results. There were direct correlations between Zn and C18:2n-6 (FDR corrected p = 0.019), total n-6 PUFAs (FDR corrected p = 0.034), and total PUFAs (FDR corrected p = 0.034). Direct correlations were found between whole blood Zn and C18:1n-9 (FDR corrected p = 0.035), C24:1n-9 (FDR corrected p = 0.023), total MUFAs (FDR corrected p = 0.023), and C18:2n-6 (FDR corrected p = 0.048) in the Cu < P50 group. In the Cu ≥ P50 group, Mg was inversely related to most FAs (All FDR corrected p < 0.05). In the Zn < P50 group, Cu was directly related to C24:1n-9, total MUFAs, C20:5n-3, C22:6n-3, total n-3 PUFAs, C20:4n-6, total n-6 PUFAs, total PUFAs, and total FAs (All FDR corrected p < 0.05). Conclusions. Whole blood Cu and Zn levels were directly linked to several FAs levels in the erythrocytes of children. The interactions of Mg, Cu, and Zn with fatty acids may affect FA metabolism, in which Mg influences FA absorption.

Minerals play important biological roles in growth, metabolism, and immunological and neurological functions [5,6]. Magnesium (Mg) acts as a necessary cofactor for several enzymes and plays a key role in glucose metabolism [7]. Iron (Fe) is associated with oxygen transport and the cytochrome enzymes reductases, catalases, and peroxidases [8]. Zinc (Zn) is required by more than 300 enzymes involved in protein synthesis [9], while copper (Cu) is important for oxidation-reduction reactions that facilitate energy production, metabolism, and detoxification [5]. Moreover, several studies have found that some mineral elements (e.g., Zn [10] and Mg [11]) are important cofactors for desaturases and elongases in endogenous FA synthesis and metabolic regulation, as well as the production of PUFAs.
Trace elements can alter FAs in human plasma, as reported in several studies. Yu et al. found that the serum levels of PUFAs are associated with the serum levels of Zn and Mg [12]. Knez et al. showed that low Zn concentrations in plasma and an altered Cu:Zn ratio are related to modified FA profiles [13]. Nevertheless, those studies did not clarify the roles of potential interactions of mineral elements in FA metabolism. Notably, the FA composition of plasma reflects recent dietary fat intake and is, thus, not a good indicator of endogenous FA production [14]. Erythrocytes better reflect long-term dietary FA intake and endogenous FA production and are widely used for this purpose [15].
The primary aim of this study was to examine the relationships between erythrocyte FA levels and whole blood calcium (Ca), Mg, Fe, Cu, and Zn concentrations, as well as the potential interactions of mineral elements in FA metabolism in Chinese children.

Study Population
This cross-sectional study was conducted in 2019 in a single primary school in Xiangtan, Hunan Province, located in South Central China. All children aged 4-7 years in this school were invited to participate. In total, 435 children were recruited, of whom 20 without collected blood samples were excluded.

Anthropometric Measurements
Anthropometric variables, including weight (kg) and height (cm), were measured using an electronic instrument. Body mass index (BMI) was calculated as weight (kg) divided by height squared (m 2 ).

Blood Sample Collection and Analysis
Blood samples were collected by venepuncture of the cubital vein using a vacuum tube with an anticoagulant. All samples were processed within 4 h of sampling, and blood samples were divided into aliquots for FA or mineral element analysis. Moreover, the lipid profile, including triglyceride (TG), cholesterol (CHOL), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) levels, were determined from serum samples on the same day of collection. A clinical chemistry analyzer AU680 (Beckman Coulter, Brea, CA, USA) and diagnostic kits (Fosun, Shanghai, China) were used to assess the plasma lipid parameters.
The whole blood samples for mineral element analysis were stored at −80 • C. The whole blood samples for FA analysis were centrifuged (3000 rpm, 1500× g, for 10 min) to separate the erythrocytes. The erythrocytes were then washed with an equal volume of saline, and the saline was removed. The remaining erythrocyte suspensions were stored in freshly butylated hydroxytoluene-treated Eppendorf vials at −80 • C.

Minerals Analysis
Whole blood Ca, Fe, Mg, Zn, and Cu concentrations were determined by flame atomic absorption spectrometry using the ContrAA 700 system (Analytik Jena GmbH, Jena, Germany), according to the method described by Guo et al. [16]. Recorded values were converted to mean values of triplicate samples, expressed in mg/L. Control serum samples with certified concentrations were used for quality control; the results obtained were in good agreement with the certified values.

Fatty Acids Determination
FA extraction and derivatization were based on a previously described procedure by Sertoglu et al. [17] with some modifications. Briefly, erythrocyte suspensions were thawed at 4 • C before the study. An internal standard working solution (IS, 100 µL), 50 µL erythrocyte suspension, and 1 mL 3 N methanolic hydrochloric acid were added to a 10 mL glass vial in the order given. The vial was closed with a screw cap to allow transmethylation to proceed at 90 • C for 1.5 h. After cooling the vials to room temperature, 2 mL hexane was added, and the vial was closed again and vortexed for 10 s. The upper (hexane) layer was transferred to a glass tube and concentrated by a drying process under a stream of nitrogen gas. Finally, the residue was dissolved in 1 mL hexane.
The samples (1 µL) were analyzed using a gas chromatography-mass spectrometer (GCMS-QP2010, Shimadzu Corp., Kyoto, Japan) and separated using an HP-88 column (dimensions, 100 m × 0.25 mm × 0.20 µm; Agilent Technologies, Santa Clara, CA, USA). The injector and detector were maintained at 230 • C. The temperature program was as follows: an initial oven temperature setting of 50 • C for 1.5 min, increase of 30 • C/min to 190 • C and holding for 5 min, followed by another increase of 8 • C/min to 230 • C and holding for 20 min. Helium was used as the carrier gas. The total analysis time was approximately 35 min. FA peaks were observed between 9.6 and 32.5 min.

Ethics Approval and Consent to Participate
All subjects provided informed consent for inclusion before participating in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of the Hunan Provincial Center for Disease Control and Prevention (HNCDC-BJ20190003).

Statistical Analysis
Continuous variables are presented as the mean (standard deviation, SD). Correlation analyses were used to explore potential relationships between different pairs of variables. For normally distributed data, Pearson's correlation analysis was used. Spearman's correlation analysis was used to analyze the data without a normal distribution. To explore the correlations between FA and mineral concentrations, partial correlation analyses were conducted after adjusting for age and sex. Blood Zn and Cu levels were categorized into two groups according to the 50th percentile (P50) to explore the potential interactions among the mineral elements. The associations of Cu and Zn levels with FA levels and their interaction effects on FAs in children were evaluated using a general linear model. In addition, the interactions of mineral elements with selected variables were examined with respect to the Zn and Cu status according to P50 stratification. All p-values of the associations or interactions with FAs or minerals were corrected for multiple testing by false discovery rate method (FDR) [20]. p < 0.05 was considered to indicate statistical significance. Statistical analyses were performed using SPSS software (version 13.0; SPSS, Inc., Chicago, IL, USA). Table 1 shows the baseline characteristics of the children. For children's sex, 52.53% was boy. The mean age and BMI were 6.02 years and 15.06 kg/m 2 , respectively. The mean TG, CHOL, HDL-C, and LDL-C levels were 1.13, 4.18, 1.55, and 2.18 mmol/L, respectively. The average whole blood Ca, Mg, Fe, Cu, and Zn concentrations were 70.37, 41.89, 448.43, 0.90, and 4.64 mg/mL, respectively.  Figure 1 shows that the main FAs in erythrocytes were saturated FAs (SFAs), accounting for more than 50% of the total FAs. C16:0 (191.36 µg/mL) had the highest concentration, followed by C18:0 (128.95 µg/mL). C18:1n-9 and C24:1n-9 were composed of more than 90% monounsaturated FAs (MUFAs). The mean concentrations of the essential PUFAs C18:3n-3 and C18:2n-6 were 1.64 and 78.29 µg/mL, respectively. The concentrations of C20:5n-3 and C22:6n-3 were 1.31 and 13.24 µg/mL, respectively, accounting for approximately 90% of the n-3 PUFAs. There were higher proportions (>90%) of C18:2n-6 and Nutrients 2022, 14, 618 5 of 13 C20:4n-6 among n-6 PUFAs. Therefore, the associations of these main FAs with minerals were the focus of our subsequent investigation. Figure 1 shows that the main FAs in erythrocytes were saturated FAs (SFAs), ac-counting for more than 50% of the total FAs. C16:0 (191.36 μg/mL) had the highest concentration, followed by C18:0 (128.95 μg/mL). C18:1n-9 and C24:1n-9 were composed of more than 90% monounsaturated FAs (MUFAs). The mean concentrations of the essential PUFAs C18:3n-3 and C18:2n-6 were 1.64 and 78.29 μg/mL, respectively. The concentrations of C20:5n-3 and C22:6n-3 were 1.31 and 13.24 μg/mL, respectively, accounting for approximately 90% of the n-3 PUFAs. There were higher proportions (>90%) of C18:2n-6 and C20:4n-6 among n-6 PUFAs. Therefore, the associations of these main FAs with minerals were the focus of our subsequent investigation.

Correlations of Mineral Element and FA Concentrations with Age, BMI, and Plasma Lipid Parameters
Correlation analyses revealed inverse relationships of the Ca level with age (FDR corrected p < 0.001) and of the Cu level with the HDL-C level (FDR corrected p = 0.011). There were direct correlations between CHOL and Zn levels (FDR corrected p = 0.044). Most of the FAs had negative associations with age (all FDR corrected p < 0.05). There were direct correlations of CHOL with C22:6n-3 levels (FDR corrected p = 0.016) and total n-3PUFAs (FDR corrected p = 0.032). No statistically relevant associations were found between BMI, TG, or LDL-C and mineral element or FA levels ( Table 2).   Table 3 lists the partial correlation coefficients adjusted for age and sex of whole blood mineral elements with erythrocyte FAs in children. There were direct correlations between Zn and C18:2n-6 (FDR corrected p = 0.019), total n-6 PUFAs (FDR corrected p = 0.034), and total PUFAs (FDR corrected p = 0.034). No statistically significant relationships were observed among the whole blood status of Ca, Mg, Fe, and Cu with any of the FAs (Table 3).   Figure 2. The correlation of different whole blood mineral elements. r: correlation coefficients. *: p < 0.05, the p-value was adjusted using false discovery rate method for multiple testing.

Indices
Ca
Furthermore, the current study showed that alterations in the erythrocyte FA composition were related to whole blood mineral element status, particularly those of Zn, which were positively associated with C18:2n-6, total n-6 PUFAs, and total PUFAs by correlation analysis. Moreover, there were weak correlations of Zn with several PUFAs, as well as between Cu and several PUFAs by general linear model after FDR correction. Several previous studies have confirmed that Zn is involved in PUFA metabolism, especially for LA:DGLA (C18:2n-6:C20:3n-6) [23][24][25]. Knez et al. reported that the plasma Cu level is directly related to oleic acid (C18:1 n-9) and alpha-linolenic acid (C18:3 n-3) [13]. Mahfouz et al. observed that Mg deficiency reduces the rate of conversion of linoleic acid (C18:2n-6) to arachidonic acid (C20:4n-6) [11]. In addition, our study found that there was a direct correlation between CHOL and Zn. Knez et al. [13] reported plasma Zn was directly and Cu:Zn ratio inversely associated with LDL-C/HDL-C and CHOL/HDL-C. On the other hand, C22:6n-3 and total n-3PUFAs had association with CHOL in this study. These results hinted that Zn may involve in lipid metabolism by altering FAs composition.
There were no interaction effects of Zn and Cu on FAs according to our general linear model. Nevertheless, our results showed the Zn level was directly related to C18:1n-9, C24:1n-9, total MUFAs, and C18:2n-6 in the Cu < P50 group, and there were weak relationships between Zn and total n-6 PUFAs, total PUFAs, and total FAs. However, no statistically significant relationships were observed between Zn and FA levels in the Cu ≥ P50 group. Additionally, the direct correlations between Cu and total MUFA, total n-3 PUFA, total n-6 PUFA, total PUFA, and total FA levels in the Zn < P50 group, whereas no statistically significant relationships were observed between Cu status and any of the FAs in the Zn ≥ P50 group. Mg was inversely related to most of the FAs in the Cu ≥ P50 groups. These results indicate that Cu may modify FA metabolism if there is a Zn deficiency, and that Zn may modify FA metabolism if there is a Cu deficiency, whereas Mg may alter FA metabolism in the presence of higher levels of Cu. Thus, there may be interactions among Cu, Zn, and Mg with respect to FA metabolism. An antagonistic relationship between Zn and Cu has been reported in some studies [26,27]. In the current study, no statistically significant relationships were observed between Cu and Zn levels in whole blood; however, positive correlations of the blood Mg level with Cu and Zn levels were found.
Our results showed that minerals are correlated with several long-chain FAs (e.g., C20:5n-3, C22:6n-3, and C20:4n-6). These FAs can be consumed from food in minor amounts; however, their status in the body mainly reflects endogenous synthesis [28]. The endogenous synthesis of these FAs depends primarily on a series of enzymes, including desaturase and elongase. Zn is a cofactor for desaturase (e.g., ∆6D and ∆5D enzymes) [4]. Cu plays an important role in the metabolic pathways of all FA desaturases, as a component of cupoprotein enzymes, resulting in increased desaturase activity [29]. In this study, Cu and Zn showed a significant positive correlation with the C18:1n-9/C18:0 ratio, representing by ∆9-D activity. Moreover, a significant positive correlation was found between the Cu level and C20:4n-6/C18:2n-6 ratio, representing desaturase/elongase activity. However, the Cu level was inversely related to the C18:0/C16:0 ratio, representing elongase activity, which may be attributed to the Fe status. Fe is also a component of a rate-limiting enzyme involved in FA synthesis [30]. A significant inverse relationship between the blood Cu and Fe level was found in this study. Therefore, the associations of FAs with Cu and Zn may involve the endogenous synthesis of FAs.
In addition, our results showed a direct correlation between C18:2n-6 and Zn levels. C18:2n-6, an essential FA for humans, must be obtained from the diet. Previous studies have reported a role of Zn in the absorption of linoleic acid, in which lower levels of Zn produce lower levels of linoleic acid [31]. Our study showed Mg was inversely related to most of the FAs (including C18:2n-6) in the Cu ≥ P50 group; however, no correlations were found between Mg and the desaturation/elongation process. These results suggest that Mg may alter the process of FA absorption.
The strengths of our study include the use of erythrocytes to reflect long-term dietary FA intake and of FA endogenous production to assess the status of individual FAs. Moreover, the potential interactions and mechanisms of mineral elements in FA metabolism was a key focus of this study. A potential limitation of the study is the cross-sectional design, which prevents clarification of the causal correlations between mineral elements and FAs. Furthermore, several factors (e.g., dietary, lifestyle, and genetic factors) [32] involved in lipid metabolism according to previous studies as potential confounding factors were not focused, which may disturb the correlation of FAs with minerals. In addition, this study included only children; thus, our findings may not be representative of the general population.

Conclusions
In this cross-sectional study, whole blood Cu and Zn levels were directly linked to several FAs levels in the erythrocytes of children. The interactions of Mg, Cu, and Zn with fatty acids may affect FA metabolism, in which Mg influences FA absorption. These results provide evidence for precision nutrition intervention in children.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/nu14030618/s1, Table S1: Partial correlation analysis of mineral elements with erythrocyte FAs in children according to P50 stratification of whole blood Cu status. Table S2: Partial correlation analysis of mineral elements with erythrocyte FAs in children according to P50 stratification of whole blood Zu status