Abstract
The aim of the present study was to examine differences and correlations in nutrient intakes and serum parameters related to nutrient intake (lipid profile, vitamins, and trace elements) in 200 lifelong Christian Orthodox Church (COC) fasters with periodic abstinence from certain foods (predominantly of animal origin) for approximately half of the year and 200 non-fasting controls, all of whom did not take dietary supplements. Nutrient intakes were assessed through three-day dietary recalls. Blood samples were drawn for the analysis of potential biomarkers of nutrient intake. Fasters had lower energy intake, due to lower fat and protein intake, compared to non-fasters (p < 0.05). Fasters also had lower intakes of vitamins A, B1, B2, B6, B12, D, folate, pantothenate, sodium, calcium, zinc, and phosphorus. Most participants (in both groups) did not meet the recommended dietary allowances of most vitamins and elements. Most serum biochemical parameters did not reflect the differences in nutrient intakes between groups, and none exhibited a correlation coefficient above 0.5 with nutrient intakes. Our findings suggest that COC fasting is associated with reduced intake of many nutrients, although this does not seem to have an impact on the blood biochemical profile.
Keywords:
biochemical profile; macronutrients; elements; nutrient intake; periodic fasting; vitamins 1. Introduction
Abstinence from certain foods, permanently or periodically, has been practiced by humans for millennia. Examples of permanent abstinence include vegetarianism (in its various nuances), whereas examples of periodic abstinence include religious fasting, as that practiced by faithful followers of the Christian Orthodox Church (COC). These abstain from certain foods, predominantly of animal origin, for periods totaling approximately half of the year []. Fasting periods are characterized by high intakes of legumes, cereals, vegetables, and fruits. As such, COC fasting constitutes a distinct dietary pattern, the study of which may offer insight into the relationship between diet and metabolic health.
We recently completed a large project examining the impact of lifelong COC fasting on nutrient intake and various health indicators. We showed that (a) men and women fully adhering to COC fasting for decades, whether being older [] or younger [], did not differ in bone mineral density, bone mineral content, or prevalence of osteopenia, osteoporosis, and bone fracture from non-fasting counterparts; (b) abstinence from dairy products and meat does not adversely affect musculoskeletal metabolism or bone density []; (c) COC fasters had lower vitamin D status than non-fasting controls, although without impact on bone health []; and (d) differences in protein intake from diverse animal and plant sources, as well as in total, had a minor (if any) impact on bone health [].
The aim of the present report is to (a) compare nutrient intakes and serum biochemical parameters between the lifelong COC fasters and non-fasters and (b) examine possible associations between these variables that might provide new information on how diet affects metabolic health.
2. Materials and Methods
2.1. Study Design and Participants
This is a cross-sectional study that included 200 fasters (131 women and 69 men) and 200 non-fasters (126 women and 74 men). The fasters had been adhering to religious fasting for a median of 15 years, ranging from 10 to 32 years and with 15 (10–26) years as the starting age of fasting. Details of the study design and the participants have been presented previously [,].
2.2. Description of COC Fasting
COC fasting involves abstaining from certain foods—predominantly of animal origin (meat, poultry, eggs, and dairy products), except seafood and snails—during five main periods, three important religious days, Wednesdays, and Fridays, totaling 159 to 197 (average, 178) days per year depending on when Easter falls and does not involve abstaining from food consumption during certain hours of the day, which characterizes intermittent fasting. The COC has set two periods, comprising a total of 47 days, of no food restriction (not even on Wednesdays or Fridays) to avoid nutrients deficiency in the body due to prolonged fasting. Between these extremes (fasting periods and periods of no food restriction) lie 17 to 23 weeks of moderate fasting, that is, only on Wednesdays and Fridays. Therefore, to make the data regarding fasters as representative of the entire year as possible, all measurements, interviews, and blood sampling were performed during these weeks.
2.3. Anthropometric Characteristics
Body weight was measured to the nearest 0.1 kg and height to the nearest 0.01 m on a digital scale with a built-in stadiometer, and body mass index (BMI) was calculated from the measurements. Waist-to-hip ratio was measured using a stretch-resistant tape.
2.4. Nutrient Intakes
Nutrient intakes were assessed through interviewer-based recalls of food consumed over three days, which included a Wednesday or Friday (during which the fasters obeyed fasting), another weekday, and a weekend day. Participants were interviewed about all foods and liquids consumed during those days and the means of the three days are presented. None of the participants in the study received any dietary supplements. Food intake records were analyzed using the Food Processor Nutrition Analysis software (ESHA, Salem, OR, USA). Vitamin and element intakes were compared to their recommended dietary allowances (RDAs) or adequate intakes (AIs), as reported by the Food and Nutrition Board of the Institute of Medicine, National Academy of Sciences [].
2.5. Biochemical Parameters
Fasting venous blood samples were drawn and treated as described []. Serum glucose, triglycerides, total cholesterol, HDL-cholesterol, LDL-cholesterol, urea, creatinine, uric acid, γ-glutamyltransferase, insulin, folic acid, vitamin B12, calcium, magnesium, iron, and phosphate were measured in two automatic analyzers, an Abbott Architect i2000SR and a Mindray BS-300, with manufacturers’ kits. The coefficients of variation for all parameters ranged from 1 to 5%, and the laboratory carrying out the analyses participated in a nationwide external quality control program.
2.6. Ethical Approval
The study was approved by the Bioethics Committee of the then Alexander Technological Educational Institute of Thessaloniki, presently International Hellenic University (31.5/5679/17-12-2013), and all procedures were in accordance with the Declaration of Helsinki. Each participant was informed about the aims, benefits, and potential risks of the study and provided written informed consent before data collection and blood sampling.
2.7. Statistical Analysis
The Kolmogorov–Smirnov test and histogram charts were used to assess normality of distribution. The distribution of almost all variables differed significantly from the normal. Thus, we report all variables as median (interquartile range) and compared groups (fasters vs. non-fasters) by using the non-parametric Mann–Whitney U test. Correlation analysis between all variables in the entire sample was performed by determining Spearman’s ρ correlation coefficient. Statistical analysis was performed using the SPSS, version 27 (SPSS, Chicago, IL, USA). All tests and corresponding p values were two-sided, and the level of statistical significance was set at α = 0.05.
3. Results
3.1. Characteristics of Participants
Fasters and non-fasters did not differ in age, body weight, height, BMI, and waist-to-hip ratio (p > 0.05). The respective values were 45 (27–58) years vs. 46 (24–57) years, 72.9 (62.9–81.4) kg vs. 71.3 (60.5–84.0) kg, 1.66 (1.60–1.72) m vs. 1.66 (1.61–1.74) m, 26.6 (23.0–29.5) kg/m2 vs. 25.6 (22.7–29.1) kg/m2, and 0.89 (0.82–0.97) vs. 0.89 (0.81–0.98).
3.2. Dietary Intakes
Total daily energy intake was lower in fasters than in non-fasters, that is, 1496 (1256–1825) kcal vs. 1611 (1303–1903) kcal in non-fasters (p = 0.027). Macronutrient intakes by the participants are presented in Table 1. The significant differences identified were the lower consumption of fat, saturated and polyunsaturated fatty acids, cholesterol, and protein by fasters compared to non-fasters.
Table 1.
Daily macronutrient intakes by fasters and non-fasters, based on three-day dietary records (median and interquartile range).
Energy distribution among macronutrients was similar between groups (Figure 1), with fat being the major contributor.
Figure 1.
Percentage energy distribution of macronutrients in fasters (A) and non-fasters (B).
Vitamin and element intakes are shown in Table 2 and Table 3, respectively. The intakes of most vitamins (A, B1, B2, B6, B12, D, folate, and pantothenate) and percentage coverage of the corresponding RDAs were significantly lower in fasters compared to non-fasters (Table 2). Most participants (in both groups) did not meet the vitamin RDAs (as can be seen by the fact that the median was below 100%), except for vitamins B1, B2, and B12 in non-fasters, as well as vitamin C in both groups.
Table 2.
Daily vitamin intakes by the two groups, based on three-day dietary records (median and interquartile range).
Table 3.
Daily element intakes by the two groups, based on three-day dietary records (median and interquartile range).
Concerning elements, fasters had significantly lower intakes and % RDAs of sodium, calcium, zinc, and phosphorus compared to non-fasters (Table 3). Only phosphorus intake by most participants in both groups was above the RDA, whereas sodium intake was above the RDA by most non-fasters.
3.3. Biochemical Profile
The values of the serum biochemical parameters measured are presented in Table 4. Fasters had significantly lower glucose, urea, vitamin B12, and phosphate, as opposed to significantly higher insulin, folate, and magnesium compared to non-fasters.
Table 4.
Serum biochemical parameters of participants in the two groups (median and interquartile range).
3.4. Correlations
A plethora of significant correlations (p < 0.05) were detected between the study parameters. To focus on the meaningful ones, we only considered correlations that were moderate or high (ρ > 0.5). Regarding correlations between the serum parameters, phosphate correlated negatively with insulin and folate (ρ = –0.68, p < 0.001 for both), whereas no correlation between serum parameters and dietary intakes had a ρ above 0.5. Additionally, there were many moderate or high correlations between dietary intakes, most of which could be explained by the concomitant abundance of certain nutrients in certain foods. For example, protein intake correlated with cholesterol intake (ρ = 0.77, p < 0.001), apparently as a result of the abundance of both in meat products, and monounsaturated fatty acids correlated with vitamin E (ρ = 0.61, p < 0.001), apparently due to the abundance of both in olive oil. Finally, all macronutrient intakes, some vitamins (vitamin B1, vitamin B2, vitamin B6), and most of the elements (Ca, Fe, Mg, P, Na, K, Zn) were strongly (ρ between 0.52 and 0.84) correlated with total energy intake.
4. Discussion
The aim of the present study was to examine whether periodic abstinence from certain foods (mainly of animal origin) for decades, according to the dictates of COC fasting, impacts nutrient intakes and the biochemical profile in relation to metabolic health. When comparing the group of fasters to that of non-fasters, a first observation was that, although the former had lower daily energy intake, they did not differ in indices of fatness (such as BMI and waist-to-hip ratio) from the latter. This could be explained by a difference in energy expenditure, although we have shown that the two groups did not differ in exercise patterns [,]. However, the possibility remains that there was a difference in daily activities that they did not perceive as exercise and, hence, did not report as such. Unfortunately, it was not possible to employ more objective means, such as activity trackers, to assess energy expenditure.
It is interesting to note that the lower energy intake by fasters was not due to lower carbohydrate intake but, rather, to lower fat and protein intakes, compared to non-fasters (Table 1). This can be explained by the fasters’ lower consumption of red meat, poultry, eggs, and dairy products, as we have shown in our previous report []. The same differences in food group consumption can explain the lower cholesterol intake by the fasters (Table 1). Nevertheless, none of the parameters of the lipidemic profile differed between groups (Table 4) or correlated with any fat intake parameter, in agreement with the tenet that dietary fat intake has little, if any, influence on the lipidemic profile [,].
A striking preponderance of fat as an energy source (nearing 50% in both groups, Figure 1) was in stark contrast to the recommendation for 20–35% of total energy []. We and others have repeatedly shown increased fat consumption by Greeks (e.g., references [,,]). Luckily, this is due in large part to high olive oil consumption, as evidenced by the dominance of monounsaturated fatty acids (27 to 28% of total energy intake). Nevertheless, energy intake from saturated fatty acids exceeded the recommended 10% [] in both groups.
Contrary to the lack of any association of the lipidemic profile with the fat intake parameters, the serum urea concentration (Table 4) did reflect the lower protein intake by the non-fasters (Table 1). This is in agreement with the acceptance of serum urea as an index of dietary protein intake [] and suggests that the former can be used as a surrogate for the latter.
Two indices of glucose homeostasis, that is, the serum glucose and insulin concentrations, exhibited opposite differences between groups. In particular, glucose was higher in non-fasters, whereas insulin was higher in fasters (Table 4). However, the values of both parameters were relatively low, and we do not see any clinical significance in their differences.
The lower energy intake by the fasters was accompanied by a lower intake of most of the vitamins and some elements, compared to non-fasters (Table 2 and Table 3). This included folate, which may seem counter-intuitive, since fasters are expected to consume more fruits and vegetables than non-fasters. However, the lower levels of folate in fasters could be explained by their lower consumption of meats (as we have shown in ref. []) since, contrary to common perception, meats are a better source of folate than fruits and vegetables [].
It is noteworthy that vitamin and element intakes were generally low, a problem that other studies have also highlighted [,,,,], although, contrary to ours, those studies did not exclude persons who consumed dietary supplements. It appears that food quality was rather poor, resulting in low nutrient density. This is in agreement with our previous finding of moderate adherence to the Mediterranean diet by both groups [,]. However, this plethora of shortfall nutrients did not seem to have an impact on most of the related biochemical parameters that we had the financial resources to measure (that is, vitamin B12, calcium, magnesium, iron, and phosphate), since most participants had values within the corresponding reference intervals. By contrast, most participants had a serum concentration of folate below its lower reference limit (3.1 ng/mL), apparently as a consequence of its low intake by both groups.
The absence of an effect of nutritional inadequacies on most of the biochemical parameters tested is in accordance with our previous findings of no effect of inadequate calcium intake on bone health in the same sample [,]. Possible reasons for the fact that some biochemical parameters do not reflect nutritional inadequacies are metabolic homeostatic mechanisms that compensate for reduced nutrient intake by increasing intestinal absorption, increasing efflux from tissues into the blood, and/or decreasing excretion.
A problem frequently encountered in nutritional studies that are based on self-reported food intake is misreporting (and usually underreporting), which is considered unavoidable []. To assess misreporting in the present study, we compared the total daily energy intake by the participants, as assessed through their dietary records, to the dietary reference values for energy of the European Food Safety Authority []. Our calculations showed a moderate underreporting of 18%. By correcting the micronutrient intakes for this underreporting, we found that the percentage coverage of the RDAs would exceed 100% in seven cases (vitamin B1, vitamin B2, vitamin B12, Na, and Fe in fasters; and vitamin B6 and Fe in non-fasters) in which it was previously below 100%. This, however, does not change the conclusion that the intakes of most micronutrients were below their RDAs.
Of interest is the negative correlation of serum phosphate with insulin found in the present study, since it agrees with the finding that, in healthy subjects, low serum phosphate was associated with reduced insulin sensitivity []. As is the case with association studies, the authors were unable to establish a cause and effect, that is, whether low phosphate is a cause or a consequence of low insulin sensitivity.
The dearth of correlations between serum biochemical parameters and nutrient intakes in our work is in accord with the few studies that have addressed the same issue or the correlation of biochemical parameters with dietary habits [,,]. Indeed, these authors have found a limited number of correlation coefficients above 0.5. It seems that nutrient metabolism is so complex and versatile that more intense research efforts are needed in the direction of discovering valid and reliable biomarkers of the composition of our diet.
5. Conclusions
Faithful followers of COC fasting had lower intakes of total daily energy, fat, and protein during periods of moderate fasting (that is, between periods of strict fasting and periods of no food restriction), periods that we consider representative of the whole year, compared to non-fasting controls. Fasters had also lower intakes of most vitamins and some elements. Most participants (in both groups) did not meet most of the vitamin and element RDAs. Most serum biochemical parameters did not reflect the differences in nutrient intakes between groups, and none exhibited a meaningful correlation with nutrient intakes.
Author Contributions
Conceptualization, N.E.R. and A.K.; methodology and data analysis, A.P., N.E.R., V.M., A.-A.K., E.V., S.K.P. and P.S.; writing—original draft preparation, A.P. and V.M.; writing—review and editing, A.P., N.E.R., V.M., M.H. and A.K.; supervision, A.K. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding. The study was funded by funds of the authors’ institutions.
Institutional Review Board Statement
The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Bioethics Committee of the then Alexander Technological Educational Institute of Thessaloniki, presently International Hellenic University; approval number 31.5/5679/17-12-2013.
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study.
Data Availability Statement
Informed consent was obtained from all subjects in the study.
Acknowledgments
We thank Metropolitan Varnavas of Neapolis and Stavroupolis Thessaloniki; Metropolitan Georgios of Kitros, Katerini and Platamon; Hieromonk, Father Luke Kipouros of Holy Trinity Monastery, Panorama of Thessaloniki; Nun, Melani Chatzipatera, Mother Superior of Assumption of Virgin Monastery, Panorama, Thessaloniki; Dimitrios Tselegidis, Professor Emeritus of the School of Theology, Aristotle University of Thessaloniki, Archimandrite Nikodemos Skrettas-Plexidas, Archpriest Athanasios Gikas, Professors of Faculty of Theology, School of Pastoral and Social Theology of Aristotle University of Thessaloniki for their help in collection of the study sample.
Conflicts of Interest
The authors declare no conflict of interest.
References
- Rodopaios, N.E.; Mougios, V.; Konstantinidou, A.; Iosifidis, S.; Koulouri, A.A.; Vasara, E.; Papadopoulou, S.K.; Skepastianos, P.; Dermitzakis, E.; Hassapidou, M.; et al. Effect of periodic abstinence from dairy products for approximately half of the year on bone health in adults following the Christian Orthodox Church fasting rules for decades. Arch. Osteoporos. 2019, 14, 68. [Google Scholar] [CrossRef] [PubMed]
- Rodopaios, N.E.; Mougios, V.; Koulouri, A.A.; Vasara, E.; Papadopoulou, S.K.; Skepastianos, P.; Dermitzakis, E.; Hassapidou, M.; Kafatos, A.G. Bone status of young adults with periodic avoidance of dairy products since childhood. Eur. J. Pediatrics 2020, 179, 645–651. [Google Scholar] [CrossRef] [PubMed]
- Rodopaios, N.E.; Manolarakis, G.E.; Koulouri, A.A.; Vasara, E.; Papadopoulou, S.K.; Skepastianos, P.; Dermitzakis, E.; Hassapidou, M.; Linardakis, M.K.; Kafatos, A.G. The significant effect on musculoskeletal metabolism and bone density of the Eastern Mediterranean Christian Orthodox Church fasting. Eur. J. Clin. Nutr. 2020, 74, 1736–1742. [Google Scholar] [CrossRef] [PubMed]
- Rodopaios, N.E.; Petridou, A.; Mougios, V.; Koulouri, A.-A.; Vasara, E.; Papadopoulou, S.K.; Skepastianos, P.; Hassapidou, M.; Kafatos, A.G. Vitamin D status, vitamin D intake, and sunlight exposure in adults adhering or not to periodic religious fasting for decades. Int. J. Food Sci. Nutr. 2021, 72, 989–996. [Google Scholar] [CrossRef] [PubMed]
- Rodopaios, N.E.; Mougios, V.; Koulouri, A.A.; Vasara, E.; Papadopoulou, S.K.; Skepastianos, P.; Hassapidou, M.; Kafatos, A.G. Dietary protein intake from different animal and plant sources plays a minor role in the bone health of adults with or without intermittent fasting for decades. Int. J. Food Sci. Nutr. 2021, 72, 704–712. [Google Scholar] [CrossRef] [PubMed]
- Nutrient Recommendations: Dietary Reference Intakes (DRI). Available online: https://ods.od.nih.gov/HealthInformation/Dietary_Reference_Intakes.aspx (accessed on 10 October 2021).
- Mozaffarian, D.; Ludwig, D.S. The 2015 US dietary guidelines: Lifting the ban on total dietary fat. JAMA—J. Am. Med. Assoc. 2015, 313, 2421–2422. [Google Scholar] [CrossRef] [PubMed]
- Dietary Guidelines Advisory Committee. Scientific Report of the 2015 Dietary Guidelines Advisory Committee: Advisory Report to the Secretary of Health and Human Services and the Secretary of Agriculture. U.S. Department of Agriculture, Agricultural Research Service, Washington, DC: 2015. Available online: https://health.gov/our-work/nutrition-physical-activity/dietary-guidelines/previous-dietary-guidelines/2015/advisory-report (accessed on 10 October 2021).
- Kabasakalis, A.; Kalitsis, K.; Tsalis, G.; Mougios, V. Imbalanced nutrition of top-level swimmers. Int. J. Sports Med. 2007, 28, 780–786. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Papadopoulou, S.K.; Barboukis, V.; Dalkiranis, A.; Hassapidou, M.; Petridou, A.; Mougios, V. Validation of a questionnaire assessing food frequency and nutritional intake in Greek adolescents. Int. J. Food Sci. Nutr. 2008, 59, 148–154. [Google Scholar] [CrossRef] [PubMed]
- Basilakis, A.; Kiprouli, K.; Mantzouranis, S.; Konstantinidis, T.; Dionisopoulou, M.; Hackl, J.M.; Balogh, D. Nutritional study in Greek-Orthodox monasteries—Effect of a 40-day religious fasting. Aktuel. Ernahrungsmed. 2002, 27, 250–255. [Google Scholar] [CrossRef]
- Wu, A. Tietz Clinical Guide to Laboratory Tests; Saunders Imprint of Elsevier: Philadelphia, PA, USA, 2016; ISBN 9781437719871. [Google Scholar]
- U.S. Department of Agriculture FoodData Central. Available online: https://fdc.nal.usda.gov/ (accessed on 1 November 2021).
- Fulgoni, V.L.; Keast, D.R.; Bailey, R.L.; Dwyer, J. Foods, Fortificants, and Supplements: Where do Americans get their nutrients? J. Nutr. 2011, 141, 1847–1854. [Google Scholar] [CrossRef] [PubMed]
- Papanikolaou, Y.; Fulgoni, V.L.I. The Role of Fortified and Enriched Refined Grains in the US Dietary Pattern: A NHANES 2009–2016 Modeling Analysis to Examine Nutrient Adequacy. Front. Nutr. 2021, 8, 655464. [Google Scholar] [CrossRef] [PubMed]
- Waem, R.V.R.; Cumming, R.G.; Blyth, F.; Naganathan, V.; Allman-Farinelli, M.; Le Couteur, D.; Simpson, S.J.; Kendig, H.; Hirani, V. Adequacy of nutritional intake among older men living in Sydney, Australia: Findings from the Concord Health and Ageing in Men Project (CHAMP). Br. J. Nutr. 2015, 114, 812–821. [Google Scholar] [CrossRef] [Green Version]
- Carr, A.C.; Lykkesfeldt, J. Discrepancies in global vitamin C recommendations: A review of RDA criteria and underlying health perspectives. Crit. Rev. Food Sci. Nutr. 2021, 61, 742–755. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- FAO. Dietary Assessment: A Resource Guide to Method Selection and Application in Low Resource Settings; FAO: Rome, Italy, 2018; ISBN 9789251306352. [Google Scholar]
- EFSA Panel on Dietetic Products, N. and A. (NDA). Scientific opinion on dietary reference values for energy. EFSA J. 2013, 11, 3005. [CrossRef] [Green Version]
- Haap, M.; Heller, E.; Thamer, C.; Tschritter, O.; Stefan, N.; Fritsche, A. Association of serum phosphate levels with glucose tolerance, insulin sensitivity and insulin secretion in non-diabetic subjects. Eur. J. Clin. Nutr. 2006, 60, 734–739. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Fraser, G.E.; Jaceldo-Siegl, K.; Henning, S.M.; Fan, J.; Knutsen, S.F.; Haddad, E.H.; Sabaté, J.; Lawrence Beeson, W.; Bennett, H. Biomarkers of dietary intake are correlated with corresponding measures from repeated dietary recalls and food-frequency questionnaires in the Adventist Health Study-2. J. Nutr. 2016, 146, 586–594. [Google Scholar] [CrossRef] [PubMed]
- Kolobarić, N.; Centner, M.G.; Šušnjara, P.; Matić, A.; Drenjančević, I. Anthropometric and biochemical parameters in relation to dietary habits as early indicator of cardiovascular impairment in young adult cohort. Int. J. Environ. Res. Public Health 2020, 17, 9208. [Google Scholar] [CrossRef] [PubMed]
- Monge-Rojas, R.; Barrantes, M.; Holst, I.; Nuñez-Rivas, H.; Alfaro, T.; Rodríguez, S.; Cunningham, L.; Cambronero, P.; Salazar, L.; Herrmann, F.H. Biochemical indicators of nutritional status and dietary intake in Costa Rican Cabécar Indian adolescents. Food Nutr. Bull. 2005, 26, 3–16. [Google Scholar] [CrossRef] [PubMed] [Green Version]
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).