TAS1R3 and UCN2 Transcript Levels in Blood Cells Are Associated With Sugary and Fatty Food Consumption in Children

Context:

New types of dietary exposure biomarkers are needed to implement effective strategies for obesity prevention in children. Of special interest are biomarkers of consumption of food rich in simple sugars and fat because their intake has been associated with obesity development. Peripheral blood cells (PBCs) represent a promising new tool for identifying novel, transcript-based biomarkers.

Objective:

This study aimed to study potential associations between the transcripts of taste receptor type 1 member 3 (TAS1R3) and urocortin II (UCN2) genes in PBCs and the frequency of sugary and fatty food consumption in children.

Design, Setting, and Participants:

Four hundred sixty-three children from the IDEFICS cohort were selected to include a similar number of boys and girls, both normal-weight and overweight, belonging to eight European countries.

Main Outcome Measures:

Anthropometric parameters (measured at baseline and in a subset of 193 children after 2 years), food consumption frequency and transcript levels of TAS1R3 and UCN2 genes in PBCs were measured.

Results:

Children with low-frequency consumption of sugary foods displayed higher TAS1R3 expression levels with respect to those with intermediate or high frequency. In turn, children with high-frequency consumption of fatty foods showed lower UCN2 expression levels with respect to those with low or intermediate frequency. Moreover, transcripts of TAS1R3 were related with body mass index and fat-mass changes after a 2-year follow-up period, with low expression levels of this gene being related with increased fat accumulation over time.

Conclusion:

The transcripts of TAS1R3 and UCN2 in PBCs may be considered potential biomarkers of consumption of sugary and fatty food, respectively, to complement data of food-intake questionnaires.

Childhood is a critical period for the establishment of dietary habits, which may affect nutrition and health status later in life (1). Individual preferences and consumption of palatable high-energy foods, such as those rich in fat or sugars, during early ages have been associated with obesity and its related complications (2–6). In this regard, different scientific advisory bodies gave specific advice for the consumption of fat and sugar (7–9). However, to establish coherent, effective food-based strategies to prevent obesity in children, more precise measures of actual intake of sugary and fatty foods are needed. In this sense, identification of biomarkers related to food consumption patterns may provide more objective measures to complement data of food intake questionnaires.

Quantification of nutrition, meaning not only what is ingested but also its biological effects, is hindered by the lack of appropriate biomarkers. At present, there are few biomarkers of food exposure, and the existing ones—with the exception of some nutrients—are generally very imprecise in measuring real intake (10). Nutrigenomic-based technologies offer new tools to address this deficiency (11). Concretely, the transcriptional profile of peripheral blood cells (PBCs), using whole blood cells or the purified fraction of peripheral mononuclear cells (PBMCs), has been proposed as a useful instrument to assess the physiological and nutritional effects of food (12, 13). Blood cells offer the advantage over other human tissues of being easily accessible from blood samples while reflecting the transcription profile occurring in other tissues (14, 15). Associations between food consumption and the transcriptional profile of PBCs have been previously reported in a microarray study, showing that gene expression profiles in PBMCs from healthy humans were different according to dietary patterns (13). There are also several studies showing changes in transcriptomic profile of PBMCs after consumption of diets rich in n3 Polyunsaturated fatty acids (16–18) and other dietary modifications [reviewed in de Mello et al (19)]. In children, expression levels of specific genes in PBCs (using whole blood cells) have been proposed as biomarkers of the metabolic status, as they are indicative of the risk of the insulin resistant or dyslipidaemic state associated with obesity (20). Gene expression in PBCs of children has also been shown to be related with the type of feeding during lactation, and may be indicative of the protective effects of breastfeeding against obesity and other metabolic alterations (21).

Here we aimed to examine potential associations between transcript levels of candidate genes in PBCs and the frequency of consumption of food rich in simple sugars and fat to identify new biomarkers of exposure to these types of food. Selected candidate genes were taste receptor type 1 member 3 (TAS1R3), for sugary food, and urocortin II (UCN2), for fatty food. The selection of genes was based on existing literature suggesting their association with sensitivity to and/or preference for sugar or fat (22, 23), and on our preliminary assays showing significant expression of these genes in PBCs. TAS1R3 gene codifies for a sweet taste receptor (TAS1R3), and single-nucleotide polymorphisms in the human TAS1R3 gene have been associated with differences in sucrose taste sensitivity (22). In turn, UCN2 encodes for urocortin II, which belongs to a family of corticotropin-releasing factor peptides with an important role in the control of food intake (23). This peptide seems to be related to the preference for high-fat food but with controversial results. Concretely, in rats, expression levels of UCN2 in hypothalamus have been correlated with the preference for high-fat diet (24), but when centrally administered, it has been shown to produce a significant decrease in the intake of high-fat diet (25). Thus we studied potential associations between transcript levels of TAS1R3 and UCN2 in PBCs and the frequency of consumption of sugary and fatty food, respectively, in a sample of children from the IDEFICS cohort (26). In addition, the predictive value of the expression of these genes on the risk of body fat accumulation over time was ascertained in a 2-year followup of a subgroup of the study population.

Subjects and Methods

Participants

Subjects involved in the study were a sample of 463 children from the IDEFICS cohort between 2 and 11 years of age. The participants were selected to include a similar number of both boys and girls, normal weight and overweight, and belonging to the eight European countries involved in the IDEFICS project as survey centers (Germany, Hungary, Italy, Cyprus, Spain, Estonia, Sweden, and Belgium). A subset of children (n= 193; 89 boys and 104 girls) was also examined after a follow-up period of 2 years.

Approval by the appropriate ethics committees was obtained by each of the eight participating centers carrying out the fieldwork. Participants were not subjected to any study procedure before both the children and their parents gave their oral (children) and written (parents) informed consent for examinations, collection of samples, subsequent analysis and storage of personal data and collected samples. Details concerning the biological samples collected and processed for the IDEFICS survey have been described elsewhere in detail (27).

Anthropometric measurements

Children in the IDEFICS surveys underwent a standardized physical examination. Anthropometric data included body weight and height, waist circumference, and the measurement of skin fold thickness. A detailed description of the anthropometric measurements adopted in the IDEFICS study, including intra- and interobserver reliability, has been published (28). The measurement of weight was carried out using an electronic scale (Tanita BC 420 SMA, Tanita Europe GmbH) to the nearest 0.1 kg with children wearing light clothes and without shoes. Height was measured using a telescopic height measuring instrument (Seca 225 stadiometer) to the nearest 0.1 cm. Body mass index (BMI) was calculated as weight divided by height squared (kg/m²). Waist circumference was measured using an inelastic tape (Seca 200), precision 0.1 cm, at the midpoint between the iliac crest and the lower coastal border or tenth rib with the subject in a standing position and recorded at the nearest 0.1 cm. Both triceps and subscapular skin fold thickness ware measured by means of a caliper (Holtain, Holtain Ltd; range, 0–40 mm). Measures were taken twice on the right hand side of the body and the mean was calculated. For the definition of overweight/obesity, children were grouped into two categories using the cut-points defined by Cole et al (29).

Dietary assessment

Food consumption frequency was estimated by the Children' s Eating Habits Questionnaire (CEHQ-FFQ) (30), in which parents or another proxy living with the child report the frequency of their child's consumption of selected food items in a typical week during the preceding 4 weeks, outside the school canteen or childcare meal provision settings. The CEHQ-FFQ asked for the consumption frequency of 43 pan-European food items of 14 food groups. Response options were as follows: “Never/less than once a week,” “1–3 times a week,” “4–6 times a week,” ”1 time per day,” ”2 times per day,” “3 times per day,” “4 or more times per day,” and “I have no idea.” Frequency categories were converted into times per week ranging from 0 to 30. The response, “I have no idea” was treated as missing. Children with more than 50% missing food items were excluded of the calculation. Consumption frequencies of simple sugar (or sugary) and fatty food were considered in the analysis. The types of foods included in both categories and the weekly frequency of consumption corresponding to the 25th and 75th percentiles are indicated in Table 1.

Table 1.

Frequencies of Consumption of Sugary and Fatty Food

Food Category	Frequency Consumption		Types of Food
	25th	75th
Sugary food	15	33	Fruit juices, sweetened drinks, sugar-added cereals, sweetened milk, sweetened yogurt, and four types of snacks, such as chocolate bars, candies, cakes, or ice cream
Fatty food	2	7	Fried potatoes, fried fish, fried meat, and fried or scrambled eggs

Expressed in times per week, corresponding to the 25th and 75th percentiles in the study population. The types of foods included in each category is indicated.

For association studies, dietary assessment data of sugary and fatty food frequencies were classified into three categories according to percentiles: low consumption (< 25th percentile), intermediate consumption (25–75th percentiles) and high consumption (> 75th percentile).

A separate 24-hour recall module was used to estimate energy and macronutrient intake. This was performed using the computer-assisted 24-hour dietary recalls, called the Self-Administered Children and Infant Nutrition Assessment (31). Parents or other caregivers as proxy respondents for recalling children's diet required information on amount (g) and type of all foods and drinks that were consumed during the previous day, starting with the first intake after waking up in the morning. Accurate estimation of portion size was assisted using standardized photographs. School meals, drinks, and snacks consumed the day prior to the 24-hour dietary recalls were assessed using a standardized observer sheet, completed by trained personnel.

Real-time qPCR analysis in whole blood cells

Real-time qPCR was used to measure mRNA expression levels as previously described (20). In short, a total of 2.5 mL of peripheral blood was collected under fasting conditions into PAXgene vacutainer tubes via antecubital fossa venipuncture, following the manufacturer's instructions (QIAGEN). Total RNA was isolated using the PAXgene blood RNA kit according to the manufacturer's instructions (QIAGEN). Primers were obtained from Sigma Genosys (Sigma-Aldrich Quimica SA). The threshold cycle (Ct) was calculated by the instrument's software (StepOne Software version 2.2.2) and the relative expression of each mRNA was calculated using the 2^−ΔΔCt method. Trim27 was used as a housekeeping gene for PCR normalization (32). The suitability of this gene as a housekeeping was also confirmed.

Statistical analysis

Statistical analysis was performed using SPSS (version 20). Data are presented as mean and the standard deviation. One-way analysis of covariance (ANCOVA) was used to assess differences between groups divided according to the percentile of distribution, adjusted for age and BMI. Bonferroni post-hoc test was used when differences were statistically significant. When indicated, ANCOVA analysis was also adjusted for food consumption frequency variables. The χ² test was used to compare proportions for categorical variables, and Student t test was used to compare two means of continuous variables. Threshold of significance was defined at P < .05.

Results

Characteristics of the population

Four hundred sixty-three children from the IDEFICS cohort were studied. Characteristics of the study population are presented in Table 2. The sample studied included boys and girls of lean/normal weight (NW) and overweight/obese (OW) in similar proportions. The mean age of all participants was 7.5 years (range, 2.3–11.6 y).

Table 2.

Characteristics of the Study Population

	All Participants (n = 463)	Boys (n = 224)	Girls (n = 239)	NW (n = 222)	OW (n = 241)
Anthropometric parameters
BMI, kg/m2	18.9 (4.2)	19.1 (4.4)	18.8 (4.0)	15.4 (1.5)	22.2 (3.1)a
Body fat, %	34.3 (9.3)	31.9 (9.6)	36.5 (8.4)	26.8 (6.3)	41.3 (5.4)a
Waist circumference, cm	62.3 (11.1)	62.9 (11.6)	61.8 (10.7)	53.5 (4.8)	70.4 (8.9)a
Sum of skin folds, mm	25.5 (12.9)	24.1 (12.9)	26.8 (12.7)	15.1 (3.9)	35.2 (10.5)a
Energy intake
kcal/d	1577 (562)	1629 (583)	1527 (538)	1602 (569)	1549 (555)
Energy from carbohydrate, %	53.0 (11.5)	52.7 (11.7)	53.3 (11.3)	52.5 (11.1)	53.5 (12.0)
Energy from fat, %	32.6 (9.0)	32.9 (9.1)	32.3 (8.9)	33.2 (9.3)	31.9 (8.7)
Energy from protein, %	15.5 (4.5)	15.3 (4.0)	15.6 (4.9)	15.3 (4.3)	15.7 (4.7)
Food intake frequency, times/wk
Sugary food	25.4 (14.5)	26.3 (15.8)	24.5 (13.2)	26.1 (14.2)	24.7 (14.9)
Fatty food	5.2 (3.8)	5.1 (3.8)	5.3 (3.9)	5.4 (4.0)	5.0 (3.7)
mRNA levels, AU
TAS1R3	81.8 (142)	84.2 (149)	81.1 (135)	80.3 (134)	84.7 (150)
UCN2	27.8 (45.1)	31.6 (44.5)	27.7 (45.6)	32.0 (46.8)	27.5 (43.4)

Abbreviation: AU, arbitrary unit.

Data are presented as Mean (SD).

^a

Differences between NW and OW (P < .05 by Student t test).

Mean BMI, percentage of body fat, waist circumference, and sum of subscapular and triceps skin folds did not differ between boy and girl participants. These variables were greater in the OW children in comparison with the NW ones (Table 2). The mean total energy intake (1577 kcal/d) and the percentage of energy from each macronutrient category, as well as the mean frequency of sugary and fatty food consumption did not differ between boys and girls nor between NW and OW children (Table 2).

Expression levels of TAS1R3 and UCN2 were not different between boys and girls. No significant differences in transcript levels of these genes were found between NW and OW children either (Table 2).

Association studies between gene expression in PBCs and food consumption frequency

Figure 1 shows the expression levels of TAS1R3 and UCN2 in PBCs in the study population classified according to percentiles of consumption frequencies of sugary and fatty food (low, intermediate and high, representing < 25th, 25–75th, and > 75th percentiles, respectively). Expression levels of TAS1R3 were associated with the frequency of consumption of food rich in simple sugars. As shown in Figure 1A, children with low-frequency consumption of food rich in simple sugars showed higher expression levels of TAS1R3 with respect to those with intermediate or high frequency.

Figure 1.

Expression levels of TAS1R3 (A) and UCN2 (B) in PBCs in the study population spread according to consumption frequencies (< 25th, 25–75th, and > 75th percentiles) of sugary food (for TAS1R3) and fatty food (for UCN2).

Results are mean ± SD expressed in arbitrary units (AUs). The number of children in each group is indicated. Statistics: a ≠ b by Bonferroni post-hoc analysis; P-values of one-way ANCOVA are indicated.

Expression levels of UCN2 were associated with the frequency of consumption of fatty food (Figure 1B). Concretely, children with high-frequency consumption of fatty food showed lower expression levels of UCN2 with respect to those with low- or intermediate-frequency consumption.

Thus, expression levels of TAS1R3 and UCN2 in PBCs seem to be related to the frequency of consumption of sugary and fatty food. To estimate their potential usefulness as biomarkers of consumption of these types of food, the same population of children was divided into three categories according to gene expression levels (low, intermediate and high, representing < 25th, 25–75th, and > 75th percentiles, respectively) and the frequency of consumption of these specific types of foods (Table 3). Significant associations were found between TAS1R3 expression levels and the frequency of consumption of sugary food (P < .01, χ² test), and between UCN2 expression levels and frequency consumption of fatty food (P < .01, χ² test). Concerning TAS1R3, most of the children with low expression levels of this gene showed intermediate or high-frequency consumption of sugary; only 16% of them showed low-frequency consumption of sugary food. Conversely, children with high expression levels showed low- or intermediate-frequency consumption of these types of food; only 20% of them showed high frequency consumption of sugary food. Intermediate levels of expression of TAS1R3 did not seem to allow a clear discrimination of the frequency of consumption of this type of food. A similar association to that of TAS1R3 with sugary food was found between UCN2 expression levels and fatty food.

Table 3.

Children With Low, Intermediate, and High Expression Levels of TAS1R3 and of UCN2 in PBCs and Spread According to Frequencies of Consumption of Sugary and Fatty Food

Expression	Frequency Consumption			χbP
	Low	Intermediate	High
Sugary food
Expression levels of TAS1R3
Low	16.3 (13)	43.8 (35)	40.0 (32)
Intermediate	20.6 (32)	45.8 (71)	23.5 (52)	.001
High	40.2 (33)	40.2 (33)	19.5 (16)
Fatty food
Expression levels of UCN2
Low	18.3 (17)	39.8 (37)	41.9 (39)
Intermediate	27.6 (54)	43.4 (85)	29.1 (57)	.00009
High	44.2 (42)	44.1 (39)	14.7 (14)

Low, < 25th percentile; intermediate, 25–75th percentile; high, > 75th percentile.

Data are presented as Percentage (n).

Association studies between gene expression in PBCs and anthropometric parameters

Considering the above-described association between expression levels of TAS1R3 and UCN2 in PBCs and the frequencies of consumption of sugary and fatty food, we next explored the potential relationship between expression levels of both genes and adiposity-related parameters in the basal study population. As shown in Table 4, no significant differences were found concerning BMI, percentage of body fat, waist circumference, or sum of triceps and subscapular skin folds between the three categories of children classified according to percentiles of expression of TAS1R3 and of UCN2. However, as expected, frequencies of sugary food consumption were different between groups classified according to TAS1R3 mRNA expression, and frequencies of fatty food consumption were different between groups classified according to UCN2 mRNA expression (Table 4).

Table 4.

Anthropometric Parameters and Food Intake Frequencies of Consumption of Sugary and Fatty Food of the Study Population Spread According to Expression Levels of TAS1R3 and of UCN2

	TAS1R3 mRNA Levels			UCN2 mRNA Levels
	Low	Intermediate	High	Low	Intermediate	High
Anthropometric parameters
BMI, kg/m2	19.3 (4.2)	18.9 (4.2)	18.3 (4.3)	19.3 (4.2)	18.6 (4.2)	19.1 (4.2)
Body fat, %	35.0 (8.8)	34.5 (9.5)	32.9 (9.2)	35.5 (9.8)	33.6 (9.2)	34.6 (8,.9)
Waist circumference, cm	63.6 (11.1)	62.3 (11.1)	60.4 (11.2)	63.0 (11.3)	61.7 (11.2)	62.4 (10.8)
Sum of skin folds, mm	27.2 (13.3)	25.5 (13.1)	23.2 (11.6)	27.5 (13.1)	24.4 (12.6)	25.5 (12.9)
Food intake frequency, times/wk
Sugary food	28.6 (15.3) a	25.7 (12.7) a	19.7 (12.8) b	24.9 (12.8)	25.6 (15.1)	24.2 (14.8)
Fatty food	5.7 (3.9)	5.4 (3.9)	4.6 (3.6)	6.1 (3.8) a	5.3 (4.0) a	4.0 (3.2) b

Low, < 25th percentile; intermediate, 25–75th percentile; high, > 75th percentile.

Data are presented as Mean (SD).

Statistics: a ÷ b by Bonferroni post-hoc analysis.

Association studies between gene expression in PBCs and anthropometric parameters in a 2-year followup

To explore the predictive value of the expression levels of TAS1R3 and UCN2 in PBCs on the risk of obesity development, the changes in anthropometric parameters (BMI, percentage of body fat, waist circumference, and sum of triceps and subscapular skin folds) over the 2-year followup were used. These data were available for a subgroup of the study population (193 children). The expression levels of TAS1R3 in PBCs of children (divided into three categories according to percentiles) were significantly associated with the 2-year variation of anthropometric parameters (Figure 2). Children with low expression levels of TAS1R3 (< 25th percentile) showed a greater increase in BMI (Figure 2A), waist circumference (Figure 2C), and sum of skin folds (Figure 2D) compared with children with expression levels greater than the 25th percentile. These associations were lost when the analyses were performed adjusting by food-consumption variables. No significant associations were found between the expression levels of UCN2 in PBCs and the above-mentioned parameters (data not shown).

Figure 2.

Changes (Δ) in anthropometric parameters—A, BMI; B, percentage of body fat; C, waist circumference; and D, sum of skin folds—over 2 years in a subset of the study population spread according to expression levels of TAS1R3 (< 25th, 25–75th, and > 75th percentiles).

Results are mean ± SD. The number of children in each group is indicated. Statistics: a ≠ b by Bonferroni post-hoc analysis; P-values of one-way ANCOVA are indicated.

Notably, no significant associations were found between the frequencies of consumption of sugary or fatty food and the 2-year variation in the anthropometric parameters mentioned above (data not shown).

Discussion

The development of new dietary-related biomarkers is crucial for nutrition research to enable a more accurate and objective assessment of food intake and to establish effective food-based strategies to prevent obesity. Analysis of PBCs transcriptomic profile is emerging as a useful tool for this purpose (13). Most of the blood cells gene expression studies have been performed in a specific subpopulation, the peripheral blood mononuclear cells (PBMCs), which include lymphocytes and monocytes and constitutes a reliable and homogeneous sample for transcriptome analysis (13, 16, 19). However, the technical procedures for the isolation of PBMCs require several methodological steps that must be strictly followed and performed immediately after blood collection to avoid ex vivo changes in gene expression profile. This may cause several logistic and technical problems, particularly when involving multicenter studies. Alternative technical procedures, such as the PAXgene blood RNA system, used in this study, allow the collection and stabilization of RNA from whole blood cells immediately upon blood sampling without the need of further manipulations (33). Thus, this procedure offers a number of technical advantages, such as the easy way of collection, storage, and transport of samples, or the reduced time of sample manipulation that make this procedure easily standardized and highly reproducible, and hence represent an attractive approach for multicenter studies (34). The limitation of procedures using whole blood cells is that they do not permit the sorting of specific cell populations given that all types of blood cells are lysed in the process. In addition, some studies have shown increased noise and reduced responsiveness with whole blood cells in comparison with PBMCs (35). Nevertheless, a significant overlap between whole blood (using PAXgene tubes) and PBMC gene expression has been demonstrated (36, 37), and hence it is expected that biomarkers identified using whole blood cells may be extended to PBMCs.

The results of the present study using whole blood cells show that the expression levels of TAS1R3 and UCN2 in PBCs of children are associated with frequencies of consumption of groups of foods potentially related to obesity development. Concretely, expression levels of TAS1R3 are associated with frequency of sugary food consumption whereas expression levels of UCN2 are associated with frequencies of fatty food consumption.

The TAS1R3 gene codifies for a sweet taste receptor (TAS1R3) that forms a heterodimer with TAS1R2 to recognize sweet-tasting molecules (38). Low expression levels of TAS1R3 in taste bud cells have been related to lower sensitivity to the sweet flavor (39), but notably, knockout mice for this receptor have a high preference for sucrose (40, 41). Thus, it seems that low sensitivity for sweet flavor may favor higher consumption of sweet-tasting food. Our findings showing that children with low-frequency consumption of sugary food display higher expression levels of TAS1R3 in PBCs in comparison with children with intermediate or high frequencies, suggest that expression levels of this gene in PBCs may be indicative of the frequency of sugary food consumption. From the present results it can be deduced that it is unlikely that children with high expression levels of this gene in PBCs (> 75th percentile) will show a frequency of consumption of sugary food above the 75th percentile (> 33 times/wk). Conversely, low expression levels of TAS1R3 (< 25th percentile) may be indicative of intermediate or high frequency of consumption of food rich in simple sugars (> 15 times/wk).

UCN2 is one of the most potent agonists of corticotropin-releasing factor 2 receptor and is implicated in food intake and anxiety-like behavior (23). It seems that UCN2 is involved in the preference for high-fat food; high expression levels of UCN2 in the hypothalamus have been found to be positively correlated with the preference for high-fat food in rats (24). However, when centrally administered, this protein has been shown to decrease high-fat diet intake, not only in lean, but also in diet-induced obese rats (25), as well as to reduce the overeating of palatable cafeteria diet (42). Notably, our results show that children with high frequency consumption of fatty food (> 75th percentile) display low expression levels of UCN2 with respect to those with low or intermediate frequencies. Therefore, analogously to the case of TAS1R3 with sugary food, low expression levels of UCN2 may be indicative of a frequency of fatty food consumption above the 25th percentile (in this case > 2 times/wk).

Thus, TAS1R3 and UCN2 expression levels in PBCs are related with the frequency of consumption of sugary and fatty food, respectively, and hence may be considered as potential biomarkers of the consumption of these types of food. In both cases, their usefulness to correctly classify individuals according to the frequency of consumption of sugary and fatty food, respectively, seems to be of particular value for low or high expression levels of this gene.

Biomarkers able to assess the excessive consumption of sugar- and fat-rich food are of high value to implement early strategies for obesity prevention because it has been described that overconsumption of both groups of foods may contribute to the obesity epidemic (43, 44). This kind of biomarker, in combination with questionnaires, may help to avoid confounding factors related to human subjective nature and deserves special interest in the case of children. Interestingly, we show here that expression levels of TAS1R3 in PBCs may predict changes in body composition in a 2-year followup. This association was lost when the analysis was adjusted by food-consumption-frequency variables, indicating that this relationship is dependent on the children's food intake. In this sense, children with low expression levels of TAS1R3 in PBCs (< 25th percentile) underwent the greatest increase in BMI, waist circumference, and skin folds compared with children with higher levels of expression. Unlike TAS1R3, expression levels of UCN2 were not significantly related to changes in the above-mentioned anthropometric parameters. In the current study, it is worth noting that no significant associations were found when data from questionnaires on sugary or fat frequency were used instead of data on gene expression in the follow-up study. This suggests that the measurement of the expression levels of TAS1R3 in PBCs might more accurately reflect the frequencies of consumption of specific types of undesirable food than the questionnaires, particularly in terms of adverse effects on body weight and fat accumulation. A limitation of the study is the fact of having fewer children in the follow-up study, as well as the lack of information on potential changes in food intake habits during this period. Additional studies in other populations would be interesting to confirm these associations.

In summary, the results of the present study show that TAS1R3 and UCN2 expression levels in PBCs are related to the frequency of consumption of sugary food (case of TAS1R3) and of fatty food (case of UCN2) in children, and hence may be considered as potential biomarkers, to be combined with data from food questionnaires. Moreover, expression levels of TAS1R3 in PBCs may predict the risk of accumulating excess fat over time more accurately than the measurement of sugary food consumption.

Acknowledgments

This work was supported by the European Community within the Sixth RTD Framework Programme Contract No. 016181 (FOOD) for the IDEFICS study and within the Seventh RTD Framework Programme Contract No. 266044 for the I.Family study. This work was also supported by the Instituto de Salud Carlos III, Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y Nutrición, CIBEROBN. The Laboratory of Molecular Biology, Nutrition and Biotechnology is a member of the European Research Network of Excellence NuGO (The European Nutrigenomics Organization, EU Contract No. FP6-506360).

This work is part of the IDEFICS study (http://www.ideficsstudy.eu/Idefics/) and the I.Family Study (http://www.ifamilystudy.eu/).

Disclosure Summary: The authors have nothing to disclose.

Abbreviations

 
• ANCOVA

  analysis of covariance

 
• BMI

  body mass index

 
• CEHQ-FFQ

  Children's Eating Habits Questionnaire

 
• NW

  normal weight

 
• OW

  overweight/obese

 
• PBC

  peripheral blood cell

 
• PBMC

  peripheral blood mononuclear cell

 
• TAS1R3

  taste receptor type 1 member 3

 
• UCN2

  urocortin II.