Children’s physiology differs from that of adults. Their higher metabolic rate, underdeveloped functional organs and relatively inefficient detoxification mechanisms make them more vulnerable to toxic compounds than adults [1
]. Clinical symptoms and health outcomes can sometimes be more severe in children; in some cases, children are even affected by some compounds while adults are not. Moreover, due to their lower body mass, risk assessment should take into account the high internal dose to body weight ratio in this population group [2
]. In addition, because of their longer life expectancy, children are prone to develop chronic syndromes in the future [3
Children’s exposure to mycotoxins is not well known and has been linked to several acute and chronic pathologies [5
]. Their symptoms and severity depend on the age, sex, immune system, and health status of the child and the specific toxins, or mixture of them, present in the diet, since ingestion is the main source of exposure to these toxic compounds [8
]. Moreover, exposure by inhalation, skin, and mucous membranes, or a combination of two or more of these routes, should also be considered [1
Acute mycotoxicoses are common in regions of developing countries in Africa and Asia, where the exposure of immunosuppressed children to certain mycotoxins is continuously high [10
]. They have been associated with poor child growth and development, recurrent infections, immune suppression, and malnutrition [20
]. Some authors suggest some relationships between impaired growth, kwashiorkor, or marasmus diseases and exposure to aflatoxins (AFs) and fumonisins (FBs) [19
]. However, others explain that the presence of AFs in the tissues of children with these diseases is due to the lower metabolization of these toxins by damaged livers [1
], and further studies are needed to clarify this aspect [25
]. In addition, high ochratoxin A (OTA) levels in children’s blood have been related to microglobulinuria due to kidney damage [1
On the other hand, the occurrence of symptoms associated with acute exposure to mycotoxins in immunocompetent children from industrialized countries has increased. In some cases, mycotoxicosis may appear without specific clinical manifestations, such as cough, nausea, vomiting, skin rash, etc., which could lead to an erroneous diagnosis. The most common acute pediatric effects of mycotoxins in developed countries are gastrointestinal diseases, when exposure to mycotoxins occurs through food or acute lung problems such as recurrent apnea or pneumonia, when exposure is due to inhalation or contact [26
Despite all of the above, levels of mycotoxins found in the diet are often low, and chronic mycotoxicosis, associated with low-level exposure, can occur worldwide. Some long-term health effects (e.g., mycotoxin-related cancers) are rarely seen in children because of their long latency periods but are of great concern because they pose a significant risk to individuals exposed to carcinogenic mycotoxins [26
]. Digestive disorders and neurological problems have also been linked to a chronic exposure to mycotoxins, but these relationships are not clear.
Mycotoxins can affect the digestive tract in two ways. The first is the alteration of the gut microbiota by exerting a toxic effect on the microbes, although this effect has been observed in studies using high concentrations of mycotoxins [27
]. In addition, mycotoxins have been described as altering the structures of the intestine. Data available in a recent review show that dietary exposure to certain mycotoxins, especially trichothecene (TCT) and patulin (PAT), cause gastrointestinal problems because they affect the intestinal barrier, impairing the permeability and integrity of epithelial cells and causing inflammation of the mucosa. In this review, authors explain that human exposure to certain mycotoxins, in particular, deoxynivalenol (DON), can be related to the etiology of chronic inflammatory bowel diseases and to the prevalence of food allergies, particularly in children [28
Other studies indicate an increased importance of diet, as the main route of mycotoxin exposure, in childhood neurobehavioral disorders [29
]. They found that OTA inhibits some genes related to autism spectrum disorder (ASD) with a gender-specific toxicity for men. However, more research needs to be done to explain the role of OTA in symptoms related to neurodevelopmental disorders [30
], and controversial data have been observed. Although De Santis et al. (2017, 2019) [29
] found a significant association between OTA in children with ASD compared to healthy children, Duringer et al. (2016) did not find this association [31
In summary, the extent and severity of children’s exposure to mycotoxins needs to be evaluated [4
]. In this assessment, biological differences between children and adults need to be taken into account. It is also necessary to increase the knowledge of the symptoms that can be caused by mycotoxins in order to detect the mycotoxicosis, apply adequate treatments, and protect the health of children [26
Exposure assessment to mycotoxins in children could be done by conducting studies on the presence of mycotoxins in food (external exposure) and by biomonitoring the presence of mycotoxins in biological samples (internal exposure).
Traditionally, the first approach has been used, although few studies have been carried out to evaluate food for children. Raiola et al. (2015) reviewed the studies conducted to understand the occurring mycotoxins in food for children. These authors concluded that more restrictive limits of mycotoxins in food for children are needed [3
]. More recently, Assunção et al. (2018) [32
] concluded that Portuguese children’s food was contaminated with several mycotoxins, especially by AFs. Another study of these authors showed that the co-occurrence of PAT and OTA in food samples in Portugal could have a major impact on intestinal health [33
]. Other authors found low levels of different mycotoxins in children’s foods [34
], but all of them suggest studies for food control, especially to reduce the exposure of children to OTA, AFs, and TCT. In general, cereal-based products are of great importance in terms of mycotoxin exposure as they are especially consumed by children [37
] and were the most contaminated by these toxic compounds, especially zearalenone (ZEA), FBs, OTA, and deoxynivalenol (DON) [34
]. Maize is considered one of the most dangerous ingredients among cereals, in particular because of the eating habits or specific pathologies of vulnerable groups, such as those with celiac disease, due to the large consumption of corn products [36
]. In a study from Tunisia, Oueslati et al. (2014) [38
] evidenced a high prevalence of several mycotoxins (including aflatoxin B1 (AFB1), OTA, and sterigmatocystin (STER)) in sorghum, a product widely consumed in the diet of children and infants. Moreover, sorghum is a potential alternative food for the celiac population, and therefore, special attention should be focused on these vulnerable groups. In Spain, exposure rates obtained for PAT and OTA, among other mycotoxins, in fruit juices, evidenced an increasing risk for children through their consumption [39
The internal exposure approach, that is, the analysis of biomarkers of exposure in biological matrices, is complementary to the analysis of mycotoxins in food and it presents some advantages: no need to identify the source of contamination, no problems related to sampling and food analysis or collection of consumption data in a varied diet and no dependence on the preparation process or the bioavailability and biological capacity of the food [37
]. Few studies have applied the internal exposure approach to assess the risk of mycotoxins to children’s health, and most of them are devoted to the exposure during the very first months or years of life [11
The current work aims to present the results obtained in the biomonitoring of mycotoxins and their conjugates in plasma from Spanish children aged 2 to 16 years. To the best of our knowledge, this is the first study in Spain evaluating the exposure to 19 mycotoxins in plasma samples from healthy children. Children with digestive problems, ASD, or attention deficit hyperactivity disorder (ADHD) were also included in the study.
Another interesting point of this study is that the obtained results can be compared with those described in a recent similar study in adult plasma samples in Spain, in which the same methodology was applied and the same compounds were analyzed [41
We present here, for the first time in Spain, the exposure of children to multiple (19) mycotoxins through the analysis of plasma samples. Plasma samples came from children aged between 2 and 16 years and with different health statuses. Forty samples were from healthy children, and 39 from children suffering from digestive disorders and ASD or ADHD disorders.
Before enzymatic treatment, OTA and OTB were detected in the plasma analysis of the children and any other mycotoxin among those analyzed—DOM-1, AFG2, AFM1, AFG1, AFB2, AFB1, ZEA, STER, T-2, HT-2, DON, FUS-X, NEO, 3-ADON, 15-ADON or DAS—was not found at levels higher than the LODs of the method in any of the samples.
According to this and other studies conducted on plasma from children and adults, OTA was the most prevalent mycotoxin and was in a concentration range of >LOD to 34.3 ng/mL in the analyzed samples. OTA was more prevalent (and with higher levels) in healthy children’s plasma than in that of sick children.
OTB appeared only in those samples from healthy children with a low incidence and always co-occurring with OTA, results very similar to those obtained in adults in Spain. This mycotoxin was not found in children with digestive disorders, ASD, or ADHD.
After enzymatic treatment, the incidence of OTA in healthy children remained high and similar to that obtained before enzymatic treatment. However, according to data obtained in adults, in a percentage of individuals’ OTA levels increased, suggesting the presence of OTA conjugates in plasma samples from healthy children. It is remarkable that, in all patient groups, the incidence and levels of OTA increased after enzymatic treatment.
STER was detected in almost all samples, but only after treatment with the enzyme mixture, similar to the way this mycotoxin was detected in adults. These results support glucuronidation as a metabolism pathway in children for this toxin. Moreover, this mycotoxin presents high incidence, especially in the three groups of sick children, for whom it was 100%.
In conclusion, it appears that the exposure of healthy adults and children to mycotoxins is similar in this region of Spain. In the case of children with digestive disorders and, also, for ASD and ADHD, the same mycotoxins have been found, but their levels, incidence and also their behavior after β-glucuronidase/arylsulfatase treatment is somewhat different. These results may indicate differences in OTA metabolism between groups of healthy children and patients.
OTA and STER should be highly considered in the risk assessment for mycotoxins. Studies concerning their sources of exposure, toxicokinetics, and the relationship between plasma levels and toxic effects are of utmost importance in both children and adults.
4. Materials and Methods
4.1. Subject Recruitment
Donors were healthy children (n = 40) and children with different pathologies (digestive disorders (n = 30), ADHD (n = 7), and ASD (n = 2)) and were selected from the Department of Pediatrics of the Clínica Universidad de Navarra. Among the patients with digestive disorders, there were 11 celiacs, 10 with fructose/lactose intolerance, 2 had eosinophilic esophagitis, 1 ulcerative colitis, 2 Helicobacter-associated gastritis, and 4 chronic abdominal pain. Written informed consent was obtained from all of them for their participation, and the procedure was approved by the Ethical Committee of the University of Navarra (project 2018.193) on 10 April 2019. All the children included in this study were under 16 years of age, thus the legal guardian/s provided the informed consent for their participation. Blood samples (n = 79) were obtained during September 2019 and March 2020. Unfortunately, the pandemic situation due to COVID-19 limited sample collection, especially for sick children. Participants only gave their gender and age as personal information.
4.2. Plasma Sample Collection
Each volunteer gave 5 mL of blood, which was collected in BD Vacutainer® Plasma Tubes (Madrid, Spain) using EDTA as an anticoagulant. Each tube was centrifuged at 12,000× g for 10 min at 4 °C, then plasma was frozen and stored at −80 °C until analysis.
4.3. Sample Analysis
The 79 samples were analyzed for mycotoxin presence before and after enzymatic treatment. Mycotoxins were analyzed using an LC system 1200 series coupled to a 6410 Triple Quadrupole (QqQ) in ESI (+) mode (Agilent Technologies, Waldbronn, Germany). The methodology was that described by Arce-López et al. (2020) [46
], and the chromatographic parameters are summed-up in Table 6
Using this methodology, 19 compounds (mycotoxins and metabolites) can be quantified. Depending on the physicochemical characteristics of the compounds, they were divided into two groups, and each one needed a different elution program for the chromatographic separation. DOM-1, AFG2, AFM1, AFG1, AFB2, AFB1, OTB, ZEA, STER, OTA, T-2, and HT-2 were included in group I. Nivalenol (NIV), DON, FUS-X, NEO, 3-ADON, 15-ADON, and DAS were included in group II.
The reagents used were as follows: deionized water (>18 MΩcm−1 resistivity) from an Ultramatic Type I system (Navarra, Spain), methanol (LC-MS grade) from Honeywell Riedel-de Haën (Seelze, Germany), acetonitrile (ACN) (HPLC grade) from Merck (Darmstadt, Germany), and formic acid (MS grade, purity >98%) and ammonium formate (MS grade) from Fluka Sigma-Aldrich (Mannheim, Germany). All mycotoxins and ochratoxin A-(phenil-d5) (OTA-d5) were obtained from Sigma-Aldrich (St. Louis, MO, USA) (reference material, purity ≥98%) as solutions in acetonitrile. Mixed stock solutions containing the mycotoxin standards (group I and II) were prepared by taking the appropriate volume from each individual standard solution, then diluted in ACN and stored at −20 °C until analysis. OTA-d5 was used in the preparation of calibrators instead of OTA. Due to their toxicity, a face shield and gloves were used when handling spiked samples.
Calibration samples were prepared by spiking human plasma. Different volumes of a mixed stock solution of mycotoxins were poured into 15 mL polypropylene centrifuge tubes and dried in an evaporator (GeneVac, SP Scientific, Ipswich, England) under vacuum at 60 °C. The residue was reconstituted using 15 µL of ACN and 450 µL of human plasma. The method for plasma treatment, before and after enzymatic treatment, was described in Arce-López et al. (2020) [46
]. Concisely, it was as follows: 0.4 mL of human plasma was passed through a Captiva EMR-lipid cartridge that contained 1.2 mL of acetonitrile (1% formic acid). The eluate was divided into two 0.4 mL portions that were evaporated until dry (60 °C). The residue resulting from one portion was reconstituted with 200 µL of 40% B-mobile phase for analyzing mycotoxins group I. The other one was reconstituted with 200 µL of 5% B-mobile phase for analyzing mycotoxins group II. The presence of Phase II metabolites in samples was assessed, reanalyzing the samples after enzymatic treatment with β-glucuronidase/arylsulfatase (from Helix Pomatia, Sigma Aldrich, Mannheim, Germany). For this purpose, 50 µL of β-glucuronidase/arylsulfatase enzyme (250 U/mL, 0.2 U/mL in phosphate buffer solution (PBS)) was added to 400 µL of plasma, and samples were maintained at 37 °C (water bath) overnight. Then, they were processed as described above.
This methodology (before and after enzymatic treatment) was successfully validated following the Food and Drug Administration (FDA) and European Medicines Agency (EMA) guidelines for bioanalytical method validation following the procedure described in Arce-López et al. (2020) [46
]. The resulting LOD values were as follows: 1.35 ng/mL for DOM-1; 0.35 ng/mL for AFG2, 0.18 ng/mL for AFM1; 0.07 ng/mL for AFG1 and AFB2; 0.04 ng/mL for AFB1; 2.70 ng/mL for HT-2; 0.40 ng/mL for OTA and OTB; 0.20 ng/mL for T-2 and STER; 1.80 ng/mL for ZEA; 9.10 ng/mL for NIV; 1.94 ng/mL for DON; 1.95 ng/mL for FUS-X; 0.18 ng/mL for NEO; 0.70 ng/mL for 3-ADON; 1.20 ng/mL for 15-ADON; and 0.15 ng/mL for DAS. Recoveries values (studied in intermediate conditions at three concentration levels) were 68.8% for STER; 77.8% for OTA-d5
; 81.6% for ZEA and NIV; 82.4% for AFB1; 82.7% for AFG1; 83.2% for OTB; 85.0% for AFB2; 88.2% for AFM1; 89.7% for DON; 89.8% for HT-2 and DOM-1; 90.1% for AFG2; 91.2% for T-2; 91.4% for FUS-X; 92.6% for 15-ADON; 93.5% for NEO; 95.1% for 3-ADON; and 97.6% for DAS (RSD ≤ 15% for all the mycotoxins).
4.4. Control of the Analytical Sequences
For analysis control, at least eight matrix-matched calibrators were analyzed along with the samples in each one of the analytical sequences. These calibrators were employed to obtain the calibration curves used for mycotoxin quantification. The criteria that should be accomplished for calibration curves were as follows: a minimum of six points, a determination coefficient (R2
) > 0.99, and a back-calculated concentration for calibration samples not different (relative error of the mean (RE) in %) by more than 15% from the nominal value (20% for LOQ level) [47
The identification of each mycotoxin in samples was carried out based on the presence of both, q and Q, product ions in the chromatogram with a ratio (q/Q in %) that did not differ more than 20% from the obtained mean ratio in calibrators of the corresponding sequence. Besides, RTs should not differ by more than 2.5% from the mean of the RTs for each mycotoxin in the calibrators [48
4.5. Statistical Analysis
Data were not normally distributed (Shapiro–Wilk test), hence equal variance was not assumed. Non-parametric tests were performed to investigate possible associations or differences between the groups. A Wilcoxon rank sum (Mann–Whitney) was used to study differences between each group (healthy and patients) and gender (boys and girls). Differences due to the enzymatic treatment (before and after treatment) were analyzed by a Wilcoxon signed-rank test.
Data above the corresponding LOD were included in the statistical analysis, whereas the LOD/2 value was used for data below LOD. All the analyses were performed using RStudio version 1.2.5019 (Boston, MA, USA). Statistical significance was set at p-value < 0.05 (95% CI).