TYPE Original Research PUBLISHED 10 October 2022 DOI 10.3389/fnut.2022.976547 Association between OPEN ACCESS ultra-processed food consumption and gut EDITED BY Mercedes G. López, Instituto Politécnico Nacional de México (CINVESTAV), Mexico REVIEWED BY microbiota in senior subjects with overweight/obesity and Tiphaine Le Roy, INSERM UMRS1269 Nutrition et obésités (nutriomique), France Sheldon George Bruno Waugh, United States Census Bureau, United States metabolic syndrome *CORRESPONDENCE Alessandro Atzeni Alessandro Atzeni1,2,3*† , María Ángeles Martínez1,2,3† , alessandro.atzeni@urv.cat Jordi Salas-Salvadó Nancy Babio1,2,3 , Prokopis Konstanti4 , jordi.salas@urv.cat Francisco J. Tinahones3,5 , Jesús Vioque6,7 , Dolores Corella3,8 , † These authors have contributed Montserrat Fitó3,9 , Josep Vidal10,11 , Isabel Moreno-Indias3,5 , equally to this work and share first authorship Salvador Pertusa-Martinez12 , Andrea Álvarez-Sala3,8 , SPECIALTY SECTION Olga Castañer3,9 , Albert Goday3,13,14 , Miguel Damas-Fuentes3,5 , This article was submitted to Nutrition and Microbes, Clara Belzer4 , Miguel Á. Martínez-Gonzalez3,15,16 , a section of the journal Frank B. Hu16,17 and Jordi Salas-Salvadó1,2,3* Frontiers in Nutrition 1 Institut d’Investigació Sanitària Pere Virgili (IISPV), Reus, Spain, 2 Universitat Rovira i Virgili, RECEIVED 23 June 2022 Department de Bioquímica i Biotecnologia, Unitat de Nutrició, Reus, Spain, 3 Centro de Investigación ACCEPTED 21 September 2022 Biomédica en Red Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Institute of Health PUBLISHED 10 October 2022 Carlos III, Madrid, Spain, 4 Laboratory of Microbiology, Wageningen University, Wageningen, CITATION Netherlands, 5 Department of Endocrinology and Nutrition, Instituto de Investigación Biomédica de Atzeni A, Martínez MÁ, Babio N, Málaga – IBIMA, Hospital Universitario Virgen de la Vic, Málaga, Spain, 6 CIBER de Epidemiología y Konstanti P, Tinahones FJ, Vioque J, Salud Pública (CIBERESP), Instituto de Salud Carlos III, Madrid, Spain, 7 Instituto de Investigación Corella D, Fitó M, Vidal J, Sanitaria y Biomédica de Alicante, Universidad Miguel Hernández (ISABIAL-UMH), Alicante, Spain, Moreno-Indias I, Pertusa-Martinez S, 8 Department of Preventive Medicine, University of Valencia, Valencia, Spain, 9 Unit of Cardiovascular Álvarez-Sala A, Castañer O, Goday A, Risk and Nutrition, Institut Hospital del Mar de Investigaciones Médicas Municipal d‘Investigació Damas-Fuentes M, Belzer C, Médica (IMIM), Barcelona, Spain, 10 CIBER Diabetes y Enfermedades Metabólicas (CIBERDEM), Martínez-Gonzalez MÁ, Hu FB and Instituto de Salud Carlos III (ISCIII), Madrid, Spain, 11 Department of Endocrinology, Institut Salas-Salvadó J (2022) Association d‘Investigacions Biomédiques August Pi Sunyer (IDIBAPS), Hospital Clinic, University of Barcelona, between ultra-processed food Barcelona, Spain, 12 Centro Salud Cabo Huertas, Alicante, Spain, 13 Medicine Department, Universitat consumption and gut microbiota in Autònoma de Barcelona, Barcelona, Spain, 14 IMIM, Hospital del Mar Medical Research Institute, senior subjects with Barcelona, Spain, 15 Department of Preventive Medicine and Public Health, University of Navarra, overweight/obesity and metabolic Pamplona, Spain, 16 Department of Nutrition, Harvard T. H. Chan School of Public Health, Boston, syndrome. Front. Nutr. 9:976547. MA, United States, 17 Channing Division for Network Medicine, Department of Medicine, Brigham doi: 10.3389/fnut.2022.976547 and Women’s Hospital and Harvard Medical School, Boston, MA, United States COPYRIGHT © 2022 Atzeni, Martínez, Babio, Konstanti, Tinahones, Vioque, Corella, The production and consumption of ultra-processed foods (UPF) has Fitó, Vidal, Moreno-Indias, Pertusa-Martinez, Álvarez-Sala, increased considerably during the last years worldwide. Collective evidence Castañer, Goday, Damas-Fuentes, shows the association between UPF consumption and adverse health Belzer, Martínez-Gonzalez, Hu and outcomes, including inflammatory gastro-intestinal disorders and obesity. The Salas-Salvadó. This is an open-access article distributed under the terms of gut microbiota has been suggested as potential mediator of the effects of the Creative Commons Attribution UPF consumption on metabolism and health. However, few studies have been License (CC BY). The use, distribution conducted in order to elucidate these aspects. Therefore, the aim of the or reproduction in other forums is permitted, provided the original present study was to assess the cross-sectional associations between UPF author(s) and the copyright owner(s) consumption and gut microbiota in a population of senior subjects (n = 645) are credited and that the original publication in this journal is cited, in within the frame of the PREDIMED-Plus trial. Eligible participants were men accordance with accepted academic and women (aged 55–75 years), without documented history of cardiovascular practice. No use, distribution or disease at enrollment, with overweight/obesity (body mass index ≤ 27 and <40 reproduction is permitted which does not comply with these terms. kg/m2 ) and metabolic syndrome. Using the information of food frequency Frontiers in Nutrition 01 frontiersin.org Atzeni et al. 10.3389/fnut.2022.976547 questionnaires, the consumption of UPF, expressed as a percentage of total dietary energy intake in kcal/day, was calculated considering those food items classified in group 4 of NOVA system. Population was categorized according to tertiles of UPF consumption. Taxonomic fecal microbiota information, along with blood biochemical parameters, anthropometric measurements and clinical data were obtained. Bioinformatics analysis was performed to study the association between fecal microbiota composition and UPF consumption. We observed that subjects allocated in the highest tertile of UPF consumption (21.4 ± 5.0 % kcal/day) presented lower adherence to MedDiet (p < 0.001) and higher total energy intake (p < 0.001). The taxonomic analysis of the fecal microbiota revealed a significant (Benjamini-Hochberg adjusted p < 0.2) positive association between specific taxa and tertiles (T) of UPF consumption: Alloprevotella (p = 0.041 vs. T2; p = 0.065 vs. T3), Negativibacillus (p = 0.096 vs. T3), Prevotella (p = 0.116 vs. T3), and Sutterella (p = 0.116 vs. T2). UPF consumption was positively associated with lower adherence to MedDiet and higher total energy intake in senior subjects with overweight obesity and metabolic syndrome. In addition, positive association with specific fecal microbiota taxa related to inflammatory gastro-intestinal diseases and low consumption of fruits and vegetables, was observed. KEYWORDS ultra-processed food, gut microbiota, mediterranean diet, 16s sequencing, overweight, obesity, metabolic syndrome Introduction juices or canned vegetables with salt), and ultra-processed foods (such as pizza or biscuits). Production and consumption of ultra-processed food (UPF) A growing body of evidence shows an association between has increased considerably during the last years worldwide, UPF consumption and an increased risk of chronic diseases displacing foods that are usually consumed in healthy eating such as obesity (5–7), metabolic syndrome (MetS) (5), diabetes patterns, abundant in unprocessed vegetables, fruits, legumes (8, 9), dyslipidemia (10), cardiovascular disease (CVD) (5, and nuts among others (1, 2). 11), cancer (12), and all-cause mortality (13, 14). Most of UPF are defined as multi-ingredient industrial formulations the studies relating UPF consumption and health outcomes manufactured from substances derived from foods and additives are cross-sectional or prospective epidemiologic studies with or ingredients not normally used in home food preparation a relatively low level of scientific evidence of cause-effect (such as hydrogenated or unesterified oils, protein isolate, relationship. However, some recent clinical studies indicate that maltodextrin, casein, and gluten), retaining little or none of their the decrease in the amount of UPF consumed should help original food characteristics. UPF include food substances not improve strategies for cardiovascular risk factors in patients used in culinary preparations, in particular flavorings, colorings, in secondary care (14, 15). These findings should serve as an sweeteners, emulsifiers and other additives used to mimic the incentive for limiting consumption of UPF, and promote natural sensory qualities of unprocessed or to mask undesirable qualities or minimally processed foods, as several national nutritional of the final product (3, 4). They are frequently highly palatable policies recommend (8, 16). and easy to eat, as they usually can be consumed at the time Considering the global increase in consumption of UPFs of purchase or with very little preparation. As a result, the and the growing evidence linking its excessive consumption UPF are commonly rich in energy density, added sugars, salt, to disease outcomes, it is important to outline the possible and saturated and trans fatty acids, and low in fiber, protein, mechanisms by which this type of food may have detrimental and micronutrients. According to the NOVA food classification health effects. One of the suggested mechanisms by which UPF system (4, 5) foods and food products can be classified into consumption may have potential effects on metabolism and four categories according to the degree of processing: low or health is through the modulation of gut microbiota composition unprocessed foods (i.e., fresh fruits, vegetables or legumes), and function (7, 17). In fact, there is growing evidence culinary ingredients (i.e., honey or sugar), processed foods (fruit demonstrating that diet may have an effect on cardiovascular Frontiers in Nutrition 02 frontiersin.org Atzeni et al. 10.3389/fnut.2022.976547 risk factors and metabolism through gut microbiota modulation. (9 possible responses ranging from never to > 6 times/day). Food processing alters the provision of macronutrient substrate Spanish food composition tables (21) were used to derive to the colon due to differing digestibility, and thereby may nutrient (sodium, saturated and trans fatty acids), fiber, alcohol impact the microbiota and its metabolic activity (17). Other (g/day), and total energy intake (kcal/day), as well as to additives and nutrient constituents have also demonstrated determine consumption of specific food groups, such as fruits to modulate the host gut microbiota after consumption and vegetables (g/day). (7, 18). However, few studies have been conducted in Food and beverage items of FFQ were classified in order to explore the relationship between UPF and gut accordance with the NOVA system proposed by Monteiro microbiota, and therefore further investigation is required in et al. (22). Foods and drinks were allocated in four different this sense. groups according to the nature, extent and purpose of The aim of the present study was to assess the baseline their industrial processing, considering a variety of physical, cross-sectional associations between UPF consumption biological, and chemical methods, as well as use of additives. and gut microbiota in a population of senior subjects For the purposes of the current study, we considered food with overweight/obesity and MetS within the frame of the items classified in the NOVA group 4 (G4), including “ultra- PREDIMED-Plus trial. processed food and drink products.” These items were further classified into subgroups: dairy products; processed meat; pre- prepared dishes, snacks and fast-foods; sweets; non-alcoholic Materials and methods beverages; alcoholic beverages (Table 1). Using the information of FFQs we calculated the consumption of UPF G4 and Study design and participants expressed as a percentage of total energy intake (in kcal/day). The proportion of UPF in the total diet was calculated for The present cross-sectional study was conducted within each participant and categorized in tertiles, corresponding to the frame of the PREDIMED (PREvención con DIeta low (first tertile), medium (second tertile), and high (third MEDiterránea)-Plus clinical trial. This trial aims to assess tertile) consumption. the long-term effects of an intensive weight loss lifestyle intervention based on an energy-restricted Mediterranean diet (MedDiet), physical activity (PA) promotion, and behavioral General assessments, anthropometric support (intervention group) on CVD events and mortality, and biochemical measurements vs. a control group following an energy-unrestricted MedDiet without any advice to increase PA (19). The trial was registered Information about sociodemographic and lifestyle aspects, at the International Standard Randomized Controlled Trial education level, individual and family medical history, and (Number: ISRCTN89898870 -data of registration: 2014-) and current medication use was collected. Leisure-time PA was approved by the institutional review board of all participating measured using the validated REGICOR questionnaire institutions. A detailed protocol is available at the web page (23) which included questions to collect information http://predimedplus.com/. Eligible participants were men and about the type of activity, frequency (number of days) women (aged 55–75 years), without documented history of and duration (min/day) performed during a representative CVD at enrollment, with overweight/obesity (body mass index month. (BMI) ≤ 27 and <40 kg/m2 ) and who met at baseline at least Waist circumference was measured midway between the three components of the MetS. lowest rib and the iliac crest using an anthropometric tape, The present study encompasses a subsample of participants body weight was measured twice using high-quality electronic from the PREDIMED-Plus recruiting centers of Reus, Barcelona calibrated scales and height was measured twice using a wall- (IMIM), Alicante, and Valencia, with available fecal microbiota mounted stadiometer. Systolic and diastolic blood pressure data at baseline (n = 656). was measured at rest 3 times using a validated semiautomatic oscillometer (Omron HEM-705CP, Kyoto, Japan) and the mean of repeated measures was used. Dietary assessments Blood samples were collected after an overnight fast. Tubes of serum and plasma were divided into aliquots, coded Dietary intake was assessed using a validated 143 item and stored at −80◦ C in a central laboratory until analyses. semiquantitative food frequency questionnaires (FFQs) (20) Plasma levels of glucose, insulin, total cholesterol, high density administered by trained dietitians. Participants reported their lipoprotein (HDL) cholesterol and triglycerides were measured average frequency and quantity of food consumed during the using standard enzymatic methods, low density lipoprotein previous year. The intake of each item was calculated by (LDL) cholesterol was calculated with the Friedewald formula multiplying a typical portion size by frequency of consumption (whenever triglycerides were <300 mg/dL). Frontiers in Nutrition 03 frontiersin.org Atzeni et al. 10.3389/fnut.2022.976547 TABLE 1 List of food items from the PREDIMED-Plus food frequency were disrupted in a 5 min lysis step using FastPrep-24TM questionnaires allocated in Group 4 (ultra-processed food and drink 5G Homogenizer (MP Biomedicals, Santa Ana, CA, USA). products) according to NOVA classification system. After extraction, the DNA was stored at −20◦ C until Dairy products Industrially-produced milkshakes further processing. DNA concentration and purity were Flavored Petit Suisse yogurt assessed with the Qubit 2.0 Fluorometer-dsDNA Broad Custard, crème caramel flan, pudding Range (BR) Assay Kit (Thermo Fisher Scientific, Waltham, Ice-cream MA, USA). The V4 region (515F-806R) of the prokaryotic 16S ribosomal Creamy cheese spreads RNA (rRNA) gene was amplified in triplicate PCR reactions. Processed meats Sandwich (deli) ham PCR was performed in a total reaction volume of 35 µl, Cured cold meats and the master mix contained 0.7 µl of a unique barcoded Pâté primer, 515F-n and 806R-n (10 µM each per reaction), 0.7 Burger µl dNTPs mixture, 0.35 µl Phusion Green Hot Start II Sweets Biscuits High-Fidelity DNA Polymerase (2 U/µl; Thermo Scientific, Whole meal biscuits Landsmeer, The Netherlands), 7 µl 5× Phusion Green HF Breakfast cereals Buffer, and 25.5 µl DNAse- RNAse-free water (Promega, Chocolate biscuits Madison, WI, USA). The amplification program included Industrially-produced confectionery 30 s of initial denaturation step at 98◦ C, followed by 25 Donuts cycles of denaturation at 98◦ C for 10 s, annealing at 50◦ C Muffins, cupcakes for 10 s, elongation at 72◦ C for 10 s, and a final extension Industrially-produced cakes step at 72◦ C for 7 min. The PCR products were visualized Churros in 1% agarose gel (∼290 bp) and purified with CleanPCR Chocolates kit (CleanNA, Alphen Aan den Rijn, The Netherlands). Soluble cocoa powder The DNA concentration of the purified PCR products was Nougat measured with Qubit dsDNA BR Assay Kit, and 200 ng from Marzipan, shortbread biscuits each purified PCR product was used for the construction Jam of the sequencing libraries, which were sequenced on an Pre-prepared dishes, Mustard Illumina Novaseq platform. To ensure the quality of the snacks and fast-foods Mayonnaise sequencing data, artificial mock communities with known Tomato sauce, Ketchup composition (24) were included in each library as positive Margarine controls, and 7 fecal samples were sequenced in duplicates. Packaged snacks Finally, to control for potential contaminant sequences, negative Croquettes control samples were included in each library, using as Instant soup DNA template, nuclease free-water or material from DNA Pizza extraction blanks. Non-alcoholic beverages Soft drinks Raw sequence data were processed using the NG-Tax No-added sugar drinks pipeline (24), with a read length of 100 nt. Paired-end Alcoholic beverages Liquors, anisette libraries were demultiplexed and only read pairs with Whisky, gin, vodka, cognac perfectly matching barcodes were used for downstream steps. Amplicon sequence variants (ASVs), were determined with the default settings and taxonomy was assigned to each ASV, using the USEARCH algorithm (25) and the Silva database Stool samples collection, DNA extraction (v138.1) (26). and 16S amplicon sequencing Stool samples were collected at home by participants and Statistical analysis kept frozen till the delivery to the laboratory. Stool samples were then separated into 250 mg aliquots and stored at −80◦ C, until The clinical characteristics of the study participants analysis. Patients exposed to antibiotics treatment (N = 6) were were described using the software IBM SPSS Statistics excluded from the present analysis. version 23 (SPSS Inc., Chicago, Illinois, USA). Numerical Microbial DNA was extracted using the QIAmp variables were considered normally distributed according PowerFecal DNA kit (Qiagen, Hilden, Germany) following to the central limit theorem and described as means and the manufacturer’s instructions. Previous extraction samples standard deviations, whereas categorical variables were Frontiers in Nutrition 04 frontiersin.org Atzeni et al. 10.3389/fnut.2022.976547 described as numbers and percentages. Differences across HDL cholesterol, LDL cholesterol, triglycerides, and glycated tertiles of UPF consumption (in % of kcal/day from total hemoglobin (HbA1c). energy intake) were tested with Pearson’s chi square test or one-way ANOVA as appropriate and p < 0.05 deemed as significant. Results Fecal microbiota data was analyzed using R (version 4.0.5) and R Studio (version 1.2.5033). Prior filtering steps, General characteristics of the study a total of 192,011,642 sequence reads for 656 samples was population obtained, clustering into 6285 ASVs. These ASVs represented 342 bacterial taxa at genus level. A cut off value of 10% of A total of 656 participants in the framework of the prevalence at genus level on the absolute abundances of the PREDIMED-Plus clinical trial with available fecal microbiota ASV counts was used to remove ASVs with a prevalence ≤10% data were included in this study. After applying quality between samples. After filtering steps, a total of 179,946,951 filtering cutoffs and excluding those participants exposed sequence reads for 645 samples and 91 genera represented were to antibiotic treatment, the study population was reduced obtained. The minimum number of sequence reads detected to 645 individuals stratified according to tertiles of UPF in the dataset was 33, the maximum number of sequence consumption. The characteristics of the study populations reads was 988,905, the average number of sequence reads are described in Table 2. A total of 215 individuals were was 278,986. allocated in tertile 1 (T1) representing those individuals The total relative abundance and percentage of each with lower consumption of UPF (7.2 ± 2.3%); 217 taxon per tertile was assessed at genus level. Principal individuals in T2 (13.1 ± 1.6%) and 213 individuals T3 component analysis (PCA) was calculated over centered (21.4 ± 5.0%), representing those individuals with higher log-ratio (CLR) transformed counts to evaluate the fecal consumption of UPF. Participants in the highest tertiles microbiota distribution of the study population among the of UPF (T2 and T3) consumption tended to be younger, tertiles of UPF consumption. Fecal microbial richness and with a higher energy consumption and lower adherence diversity was assessed by calculating Chao1, Shannon and to MedDiet. Simpson indices on absolute abundance counts and differences From the total energy of UPF consumed, the across tertiles of UPF consumption tested with one-way main contributors (Figure 1) were represented by non- ANOVA. Beta diversity was calculated in terms of Euclidean alcoholic beverages (26%), sweets (26%), processed distance over CLR-transformed counts and permutational meats (21%), preprepared dishes, snacks and fast- multivariate analysis of variance (PERMANOVA) performed foods (14%), dairy products (11%) and alcoholic to test differences in microbiota dissimilarity across tertiles beverages (2%). of UPF consumption. The log-normalized Firmicutes-to- Bacteroidetes ratio was computed based on the relative abundance between the phylum Firmicutes and Bacteroidetes Fecal microbiota diversity and and differences across tertiles of UPF consumption tested using dissimilarity and main phyla distribution one-way ANOVA. across study groups The association between tertiles of UPF consumption and fecal microbiota was assessed using the R package No statistically significant differences in alpha diversity MaAsLin2 (27) (version 1.2.0). Specifically, counts were indices calculated, Chao1, Shannon, and Simpson, across CLR-transformed, and model adjusted for different potential tertiles of UPF consumption (Supplementary Figure 1 confounders (age, sex, recruiting center, smoking habits, and Supplementary Table 1) were observed. PCA over diabetes prevalence, BMI categories (overweight, with BMI the CLR-transformed taxonomic feature counts shown 27–29.9 kg/m2 ; obese, with BMI 30–39.9 kg/m2 ; severe that PC1 and PC2 account ∼10.15% and 8% to the obese, with BMI ≥40 kg/m2 ), and tertiles of PA. Multiple total variation, respectively. PCA plot showed that testing correction was performed using Benjamini-Hochberg the fecal microbiota samples did not cluster based on procedure, and p < 0.2 deemed as significant. Spearman’s tertiles of UPF consumption (Supplementary Figure 2). correlation was used to test the relationship between taxa Furthermore, the result of the PERMANOVA test based significantly associated with tertiles of UPF consumption, on Euclidean distance did not show statistically significant different G4 UPF items categories and cardiovascular risk differences across tertiles of percentage in UPF energy factors: body weight, waist circumference, and body mass index intake (Supplementary Table 2). There were no differences (BMI), systolic blood pressure (SBP), diastolic blood pressure in Firmicutes-to-Bacteroidetes ratio across tertiles UPF (DBP), fasting plasma glucose (FPG), insulin, total cholesterol, consumption (Supplementary Table 3). Frontiers in Nutrition 05 frontiersin.org Atzeni et al. 10.3389/fnut.2022.976547 TABLE 2 Characteristics of the study population according to tertiles of UPF consumption. Tertiles of UPF T1 (N = 215) T2 (N = 217) T3 (N = 213) p value& consumption (% 7.177 ± 2.349 13.094 ± 1.572 21.434 ± 4.956 kcal/day from total energy intake) Women 103 (47.9) 99 (45.6) 102 (47.9) 0.861 Recruiting center 0.048 Alicante 51 (23.7) 49 (22.6) 45 (20.7) Barcelona IMIM 26 (12.1) 21 (9.7) 19 (8.9) Reus 119 (55.3) 124 (57.1) 110 (51.6) Valencia 19 (8.8) 23 (10.6) 41 (19.2) Smoking habits 0.659 Ex-smoker 81 (37.7) 87 (40.1) 84 (39.4) Never smoked 104 (48.4) 108 (49.8) 98 (46.0) Smoker 30 (14.0) 22 (10.1) 31 (14.6) Education 0.726 NA 1 (0.5) 1 (0.5) 0 (0.0) Higher education 40 (18.6) 45 (20.7) 32 (15.0) Primary school 113 (52.6) 109 (50.2) 119 (55.9) Secundary school 61 (28.4) 62 (28.6) 62 (29.1) Civil status 0.724 NA 0 (0.0) 0 (0.0) 1 (0.5) Divorced 14 (6.5) 10 (4.6) 17 (8.0) Married 161 (74.9) 174 (80.2) 160 (75.1) Separated 7 (3.3) 7 (3.2) 6 (2.8) Single 8 (3.7) 9 (4.1) 11 (5.2) Widowed 25 (11.6) 17 (7.8) 18 (8.5) BMI category 0.738 Overweight (27–29.9 kg/m2 ) 54 (25.1) 55 (25.3) 49 (23.0) Obese (30–39.9 kg/m2 ) 102 (47.4) 95 (43.8) 107 (50.2) Severe obese (≥ 40 kg/m2 ) 59 (27.4) 67 (30.9) 57 (26.8) Hypertension 204 (94.0) 206 (94.9) 198 (93.0) 0.692 Diabetes 60 (27.9) 48 (22.1) 53 (24.9) 0.381 Hypercholesterolemia 146 (67.9) 142 (65.4) 139 (65.3) 0.810 Age (years) 65.7 ± 4.9 64.8 ± 4.9* 64.5 ± 4.8* 0.023 Body weight (kg) 86.8 ± 13.1 87.2 ± 13.3 87.1 ± 12.9 0.943 Waist circumference (cm) 107.8 ± 10.5 107.7 ± 9.7 107.5 ± 9.9 0.962 BMI (kg/m2 ) 32.8 ± 3.5 32.9 ± 3.6 33.0 ± 3.5 0.906 FPG (mg/dL) 116.7 ± 22.2 116.9 ± 28.4 112.3 ± 21.6 0.087 Insulin (mU/mL) 19.8 ± 12.2 19.9 ± 11.7 19.2 ± 9.7 0.774 Total cholesterol (mg/dL) 200.4 ± 37.9 199.8 ± 37.9 201.4 ± 35.8 0.900 HDL cholesterol (mg/dL) 49.4 ± 10.4 49.9 ± 11.9 49.3 ± 12.1 0.853 LDL cholesterol (mg/dL) 122.2 ± 34.1 117.6 ± 30.8 121.8 ± 31.0 0.268 Triglycerides (mg/dL) 152.8 ± 74.0 164.0 ± 93.2 167.4 ± 111.9 0.245 Glycated hemoglobin (%) 6.1 ± 0.8 6.1 ± 0.9 6.0 ± 0.8 0.668 UPF energy intake (kcal/day) 171.6 ± 70.6 329.6 ± 77.3* 578.0 ± 234.3*† < 0.001 Total energy intake (kcal/day) 2363.4 ± 482.1 2516.1 ± 499.1* 2677.2 ± 482.1*† < 0.001 Adherence to MedDiet score 8.7 ± 2.4 8.1 ± 2.6* 7.7 ± 2.3* < 0.001 UPF, ultra-processed food; NA, not available, BMI, body mass index; FPG, fasting plasma glucose; HDL, high density lipoprotein; HDL, low density lipoprotein. Data expressed in number (percentage) for categorical variables and mean ± standard deviation for numerical variables. & Differences across tertiles calculated with Pearson’s chi square or one-way ANOVA, differences between tertiles calculated with Pearson’s chi square or Student’s t-test. *p < 0.05 vs. T1; †p < 0.05 vs. T2. Frontiers in Nutrition 06 frontiersin.org Atzeni et al. 10.3389/fnut.2022.976547 cardio-metabolic risk (30), suggesting that a high frequency of UPF consumption could be related to obesity and several cardiometabolic alterations. In the present study, the population was categorized according to tertiles of UPF consumption, and we observed that subjects allocated in the highest tertile presented lower adherence to MedDiet and higher total energy intake. These results were in line with the aforementioned articles, and very similar to the ones obtained previously by Cuevas-Sierra et al. where higher values of total energy intake and a lower adherence FIGURE 1 to MedDiet were observed in subjects consuming more than 5 Pie chart showing the proportion (in % kcal/day) of UPF UPF servings/day (18). consumption (group 4 of the NOVA classification system) in the total diet of the study population. According to our findings, we did not observe any association between alpha and beta diversity nor with Firmicutes-to-Bacteroidetes ratio in our study population. In relation to that, it has been reported that food processing may Association between specific fecal disturb gut microbiota composition also reducing bacterial microbiota taxa and UPF consumption diversity. However, the mechanisms by which UPF affects gut microbial homeostasis remain unclear (31). In addition, The analysis performed with MaAsLin2 shown a positive there are studies linking a varied and omnivorous diet, and association between fecal microbiota genera Alloprevotella and consequently higher adherence to MedDiet, with an increase Sutterella, representing, respectively, 1.2% and 0.6% of the in the ratio Firmicutes-to-Bacteroidetes (32, 33). In line with total relative abundance (Supplementary Table 4), and T2 of these studies, Partula et al. aimed to explore the association percentage in UPF energy intake, and a positive association between usual diet and gut microbiota composition, established between fecal microbiota genera Alloprevotella, Negativibacillus, through 16S rRNA gene sequencing in stool samples from and Prevotella, representing respectively 0.3%, 0.03% and 0.3% healthy French adults (34). They observed that food items of the total relative abundance (Supplementary Table 4), and T3 for which a limited consumption is generally recommended of percentage in UPF energy intake (Table 3). (fried products, sodas or sugary drinks, fatty sweet products, In addition, the results of the Spearman correlation did processed meats, ready-cooked meals, desserts, and cheese) not show any relevant association between taxa significantly were negatively associated with alpha-diversity and contributed associated with UPF consumption and UPF items categories to shifts within microbiota composition (beta-diversity). These (Supplementary Figure 3 and Supplementary Table 5) nor contradictory results in relation to the aforementioned studies with cardiovascular risk factors (Supplementary Figure 4 and can be explained by a lower consumption of UPF in our senior Supplementary Table 6). study population. In the present study we observed a positive association between 4 different taxa abundances and higher UPF Discussion consumption. Specifically, we found that abundance of genera Alloprevotella, Negativibacillus, Prevotella and Sutterella, The current cross-sectional study aimed to assess the was positively associated with the highest consumption of association between UPF consumption and fecal microbiota UPF. These positive associations are in accordance with other composition in senior subjects with overweight/obesity and recently published studies. The association between specific MetS within the frame of the PREDIMED-Plus trial. A previous food items and relative abundances of taxa tested by Partula study conducted in the whole PREDIMED-Plus cohort shown et al. shown that Alloprevotella was negatively associated that higher consumption of UPF was associated with higher with the consumption of fruits (34). Alloprevotella was found visceral and overall adiposity accumulation (28). Similar results enriched in the gut microbial community determined by have also been reported some years before by Mendoça et al., in absolute quantification 16S rRNA gene amplicon sequencing of the context of a prospective cohort study in healthy middle-aged healthy individuals exposed to 4 days dietary fiber intervention Spanish individuals, in which those participants in the highest (35). However, these findings should be interpreted with caution quartile of UPF consumption showed an increased risk of as obtained in a small sample size. developing overweight or obesity (29). A recent comprehensive Negativibacillus has been recently discovered in the systematic review explored extant relevant nutrition literature microbial community of human stool samples and it has reporting increasing evidence of UPF exposure associated been found enriched in patients with ulcerative colitis (UC) with adverse health outcomes such as overweight, obesity and (36). There is not much information on this genus in the Frontiers in Nutrition 07 frontiersin.org Atzeni et al. 10.3389/fnut.2022.976547 TABLE 3 Results of the association analysis performed with MaAsLin2 between taxonomic feature counts at genus level and tertiles of percentage in UPF energy intake. Feature Tertile# coef stderr pval padj Alloprevotella 2 0.430 0.133 0.001 0.041 3 0.410 0.134 0.002 0.065 Negativibacillus 2 0.071 0.058 0.224 0.725 3 0.170 0.059 0.004 0.096 Prevotella 2 0.161 0.119 0.178 0.687 3 0.335 0.121 0.006 0.116 Sutterella 2 0.330 0.119 0.006 0.116 3 0.093 0.120 0.441 0.841 Association tested with GLM adjusted for age, sex, recruiting center, smoking habits, diabetes prevalence, BMI categories, tertiles of physical activity. Multiple testing correction was performed using Benjamini-Hochberg procedure and p < 0.2 deemed as significant. # (reference: Tertile 1). Significant associations are indicated in bold. bibliography, however, in a previous published study it has been Our study has some limitations that deserve to be discussed. found that the abundance of Negativibacillus was positively First, the observational design of the study did not make associated with obesity in mice (37). possible to conclude on causality or directionality as these results Prevotella species, as Negativibacillus, have been associated indicate just associations. However, observational design is the with patients with UC (36). This genus has also been associated most indicated approach to explore these type of associations, with plant-based diet, rich in fiber and carbohydrates, and has as longitudinal studies based on interventions including UPF been considered as a biomarker of health. Nevertheless, there intake may be detrimental for health and could cause ethical are important confounding factors such as microbial exposure, issues. Secondly, the results cannot be extrapolated to other lifestyle and geography that may affect this associative evidence populations since participants included in the analysis were (38). In addition, the potential pathogenic role of certain senior Mediterranean subjects with overweight/obesity and Prevotella strains has been also highlighted (39). The strain- MetS. Under the methodological point of view, the nature of level diversity may explain the controversial role of Prevotella 16S sequencing limits the taxonomic profiling to genus-level species in health and disease (40). This could partially explain resolution. Finally, the assessment of food intake through a FFQ the association that we found between this taxon and higher UPF is sensitive to possible measurements errors. Nevertheless, food- consumption, as the response to different dietary patterns may based FFQs have been widely used as a tool in epidemiological be strain dependent and undetectable due to the next generation studies since the 1990’s (52). sequencing techniques utilized that limited us to genus level On the other hand, the current study also has some resolution, as also emphasized by De Filippis et al. (40). strengths, such as the large cohort sample size, the extensive data Finally, Sutterella, has been recently linked to inflammatory collection, and the use of several potential confounding factors gastrointestinal diseases (41–44). The relationship between which minimizes the possibility of reverse causality bias. Finally, diet and development of inflammatory gastrointestinal tract it is also important to highlight that the study setup provides an disorders has been indicated in several epidemiological ideal scenario to explore specific gut microbial profiles related to studies (45). Previous meta-analysis studies showed dietary patterns. that soft drink consumption and sucrose intake were associated with 69% and 10% increased risk of UC development, respectively (46, 47). In addition, consumption of fruits and vegetables was related to decreased UC Conclusions development (48). This could in part explain the positive association between Sutterella and UPF consumption in In conclusion, the results of this present observational our cohort. study conducted within the frame of the PREDIMED-Plus Despite these findings, in the current study we did not cohort revealed that the consumption of UPF (rich in energy observe any correlation between the above-mentioned UPF- density, added sugars, salt, and saturated and trans fatty acids, associated taxa, UPF items categories and CVD risk factors, and low in fiber, protein, and micronutrients) was positively respectively. Nevertheless, lower adherence to MedDiet was associated with lower adherence to MedDiet and higher total observed in those participants in the highest tertile of UPF energy intake. In addition, we observed a positive association energy consumption. These results were in consonance with with specific taxa (Alloprevotella, Negativibacillus, Prevotella and previous studies (49–51). Sutterella) related to inflammatory gastro-intestinal diseases and Frontiers in Nutrition 08 frontiersin.org Atzeni et al. 10.3389/fnut.2022.976547 low consumption of fruits and vegetables, among others, across MM, and JS-S drafted the manuscript. All authors reviewed the tertiles of UPF consumption. manuscript for important intellectual content and approved the Diet and nutritional status are important determinants final version to be published. in human health. The role of diet in shaping the gut microbiota could be key to improve host’s health. Consequently, further future studies including whole metagenome shotgun Funding sequencing would be crucial to obtain more detailed taxonomic identification. Furthermore, information generated from This work was supported by the official Spanish Institutions metabolomics analysis would be useful to improve and for funding scientific biomedical research, CIBER Fisiopatología strengthen the results. In addition, more in-depth studies de la Obesidad y Nutrición (CIBEROBN) and Instituto de Salud regarding the association between taxa and UPF consumption Carlos III (ISCIII), through the Fondo de Investigación para are needed in order to further reduce the controversies between la Salud (FIS), which is co-funded by the European Regional the authors. The detection of unhealthy dietary patterns and Development Fund (three coordinated FIS projects leaded by association with gut microbial profiles would be essential for JS-S, including the following projects: PI13/00462, PI16/00501 preventing related chronic diseases and to improve public and PI19/00576); The Generalitat Valenciana PROMETEO health strategies. 17/2017, PROMETEO 21/2021 and APOSTD/2020/164 (to AÁ-S). This research was also partially funded by EU- H2020 Grant Eat2beNICE/H2020-SFS-2016-2 and by NIH grant Data availability statement R01DK127601. None of the funding sources took part in the design, collection, analysis, interpretation of the data, or writing The datasets generated and analyzed during the current the report, or in the decision to submit the manuscript for study are not publicly available due to data regulations publication. Food companies Hojiblanca (Lucena, Spain) and and for ethical reasons, considering that this information Patrimonio Comunal Olivarero (Madrid, Spain) donated extra might compromise research participants’ consent because our virgin olive oil; and the Almond Board of California (Modesto, participants only gave their consent for the use of their data CA), American Pistachio Growers (Fresno, CA), and Paramount by the original team of investigators. However, collaboration Farms (Wonderful Company, LLC, Los Angeles, CA) donated for data analyses can be requested by sending a letter to nuts for the PREDIMED-Pilot study. AA was supported by the PREDIMED-Plus steering Committee (predimed_plus_ the European Union’s Horizon 2020 research and innovation scommittee@googlegroups.com). The request will then be programme under the Marie Skłodowska-Curie grant agreement passed to all the members of the PREDIMED-Plus Steering No. 713679 and from the Universitat Rovira i Virgili (URV). Committee for deliberation. MM was supported by Sara Borrell -CD21/00045- Instituto de Salud Carlos III (ISCIII). IM-I was supported by a Miguel Ethics statement Servet II contract -CPII21-00013- Instituto de Salud Carlos III (ISCIII). The studies involving human participants were reviewed and approved by the study was conducted according to the guidelines of the Declaration of Helsinki, and approved by Acknowledgments the Hospital Universitari Sant Joan de Reus Ethics Committee (protocol code 13-07-25/7proj2, date of approval: 25/07/2013), We thank all the volunteers for the participation and and the Comité de Ética de la Investigación Provincialde Málaga personnel for the contribution in the PREDIMED-Plus trial. (protocol code Predimed+DM/01, date of approval: 27/11/ We also thank all the investigators of the PREDIMED-Plus 2014). The patients/participants provided their written informed study. CIBEROBN (Centros de Investigación Biomédica en consent to participate in this study. The patients/participants Red: Obesidad y Nutrición) is an initiative of ISCIII, Madrid, provided their written informed consent to participate in Spain. The authors also thank the PREDIMED-Plus Biobank this study. Network as a part of the National Biobank Platform of the ISCIII for storing and managing the PREDIMED-Plus biological samples. JS-S senior author of this study gratefully acknowledges Author contributions the financial support by ICREA under the ICREA Academia programme. The research groups want to thank Jasper Koehorst All the principal PREDIMED-Plus investigators from the and Bart Nijsse from the Unlock platform for their support in respective centers contributed to study concept and design and processing the data with the NG-tax pipeline. We also want to to data extraction from the participants (FT, JVio, DC, MF, JVid, thank Athanasia Ioannou for her work with the stool samples and JS-S). AA and MM performed the statistical analyses. AA, preparation. Frontiers in Nutrition 09 frontiersin.org Atzeni et al. 10.3389/fnut.2022.976547 Conflict of interest organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or The authors declare that the research was conducted in the claim that may be made by its manufacturer, is not guaranteed absence of any commercial or financial relationships that could or endorsed by the publisher. be construed as a potential conflict of interest. Supplementary material Publisher’s note The Supplementary Material for this article can be All claims expressed in this article are solely those of the found online at: https://www.frontiersin.org/articles/10.3389/ authors and do not necessarily represent those of their affiliated fnut.2022.976547/full#supplementary-material References 1. Salas-Salvadó J, Martinez-González MÁ, Bulló M, Ros E. The role of mortality among middle-aged adults in France. JAMA Intern Med. (2019) diet in the prevention of type 2 diabetes. Nutr Metab Cardiovasc Dis. (2011) 179:490. doi: 10.1001/jamainternmed.2018.7289 21:09. doi: 10.1016/j.numecd.2011.03.009 15. Beslay M, Srour B, Méjean C, Allès B, Fiolet T, Debras C, et al. Ultra-processed 2. Martínez Leo EE, Segura Campos MR. Effect of ultra-processed diet on food intake in association with BMI change and risk of overweight and obesity: A gut microbiota and thus its role in neurodegenerative diseases. Nutrition. (2020) prospective analysis of the French NutriNet-Santé cohort. Souza Lopes AC, editor. 71:110609. doi: 10.1016/j.nut.2019.110609 PLOS Med. (2020) 17:e1003256. doi: 10.1371/journal.pmed.1003256 3. Steele EM, Baraldi LG, Da Costa Louzada ML, Moubarac JC, Mozaffarian D, 16. Bonaccio M, Di Castelnuovo A, Costanzo S, De Curtis A, Persichillo M, Sofi Monteiro CA. Ultra-processed foods and added sugars in the US diet: Evidence F, et al. Ultra-processed food consumption is associated with increased risk of all- from a nationally representative cross-sectional study. BMJ Open. (2016) 6:1– cause and cardiovascular mortality in the Moli-sani Study. Am J Clin Nutr. (2021) 8. doi: 10.1136/bmjopen-2015-009892 113:446–55. doi: 10.1093/ajcn/nqaa299 4. Monteiro CA, Cannon G, Moubarac J-C, Levy RB, Louzada MLC, 17. Muralidharan J, Moreno-Indias I, Bulló M, Lopez JV, Corella D, Castañer Jaime PC. The UN decade of nutrition, the NOVA food classification O, et al. Effect on gut microbiota of a 1-y lifestyle intervention with and the trouble with ultra-processing. Public Health Nutr. (2018) 21:5– Mediterranean diet compared with energy-reduced Mediterranean diet and 17. doi: 10.1017/S1368980017000234 physical activity promotion: PREDIMED-Plus Study. Am J Clin Nutr. (2021) 114:1148–58. doi: 10.1093/ajcn/nqab150 5. Martinez-Perez C, San-Cristobal R, Guallar-Castillon P, Martínez-González MÁ, Salas-Salvadó J, Corella D, et al. Use of different food classification 18. Cuevas-Sierra A, Milagro FI, Aranaz P, Martínez JA, Riezu-Boj JI. Gut systems to assess the association between ultra-processed food consumption microbiota differences according to ultra-processed food consumption in a Spanish and cardiometabolic health in an elderly population with metabolic syndrome population. Nutrients. (2021) 13:1–20. doi: 10.3390/nu13082710 (PREDIMED-Plus Cohort). Nutrients. (2021) 13:2471. doi: 10.3390/nu13072471 19. Salas-Salvadó J, Díaz-López A, Ruiz-Canela M, Basora J, Fitó M, Corella 6. Nardocci M, Leclerc B-S, Louzada M-L, Monteiro CA, Batal M, Moubarac J-C. D, et al. Effect of a lifestyle intervention program with energy-restricted Consumption of ultra-processed foods and obesity in Canada. Can J Public Heal. mediterranean diet and exercise on weight loss and cardiovascular risk factors: (2019) 110:4–14. doi: 10.17269/s41997-018-0130-x one-year results of the PREDIMED-Plus trial. Diabetes Care. (2018) 42:dc180836. doi: 10.2337/dc18-0836 7. Cuevas-Sierra A, Ramos-Lopez O, Riezu-Boj JI, Milagro FI, Martinez JA. Diet, gut microbiota, and obesity: links with host genetics and epigenetics and potential 20. Fernández-Ballart JD, Piñol JL, Zazpe I, Corella D, Carrasco P, Toledo applications. Adv Nutr. (2019) 10:S17–30. doi: 10.1093/advances/nmy078 E, et al. Relative validity of a semi-quantitative food-frequency questionnaire in an elderly Mediterranean population of Spain. Br J Nutr. (2010) 103:1808– 8. Srour B, Fezeu LK, Kesse-Guyot E, Allès B, Debras C, Druesne-Pecollo 16. doi: 10.1017/S0007114509993837 N, et al. Ultraprocessed food consumption and risk of type 2 diabetes among participants of the nutrinet-santé prospective cohort. JAMA Intern Med. (2020) 21. Moreiras O, Carbajal Á, Cabrera L, Cuadrado C. Tablas de composición de 180:283. doi: 10.1001/jamainternmed.2019.5942 alimentos. 19a edició. Pirámide (2018). 9. Llavero-Valero M, Escalada-San Martín J, Martínez-González MA, Basterra- 22. Monteiro CA, Cannon G, Jaime PC, Canella DS, Louzada ML, Calixto G, Gortari FJ, de la Fuente-Arrillaga C, Bes-Rastrollo M. Ultra-processed foods and et al. Food Classification. Public Health NOVA. The Star Shines Bright. (2016). type-2 diabetes risk in the SUN project: A prospective cohort study. Clin Nutr. 23. Molina L, Sarmiento M, Peñafiel J, Donaire D, Garcia-Aymerich (2021) 40:2817–24. doi: 10.1016/j.clnu.2021.03.039 J, Gomez M, et al. Validation of the regicor short physical activity 10. Donat-Vargas C, Sandoval-Insausti H, Rey-García J, Moreno-Franco B, questionnaire for the adult population. Lucía A, editor. PLoS ONE. (2017) Åkesson A, Banegas JR, et al. High consumption of ultra-processed food is 12:e0168148. doi: 10.1371/journal.pone.0168148 associated with incident dyslipidemia: a prospective study of older adults. J Nutr. 24. Ramiro-Garcia J, Hermes GDA, Giatsis C, Sipkema D, Zoetendal EG, (2021) 151:2390–8. doi: 10.1093/jn/nxab118 Schaap PJ, et al. NG-Tax, a highly accurate and validated pipeline for 11. Srour B, Fezeu LK, Kesse-Guyot E, Allès B, Méjean C, Andrianasolo RM, et al. analysis of 16S rRNA amplicons from complex biomes. F1000Research. (2018) Ultra-processed food intake and risk of cardiovascular disease: prospective cohort 5:1791. doi: 10.12688/f1000research.9227.2 study (NutriNet-Santé). BMJ. (2019) 2019:1451. doi: 10.1136/bmj.l1451 25. Edgar RC. Search and clustering orders of magnitude faster than BLAST. 12. Fiolet T, Srour B, Sellem L, Kesse-Guyot E, Allès B, Méjean C, Bioinformatics. (2010) 26:2460–1. doi: 10.1093/bioinformatics/btq461 et al. Consumption of ultra-processed foods and cancer risk: results from 26. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA NutriNet-Santé prospective cohort. BMJ. (2018) 2018:k322. doi: 10.1136/bmj. ribosomal RNA gene database project: improved data processing and web-based k322 tools. Nucleic Acids Res. (2012) 41:D590–6. doi: 10.1093/nar/gks1219 13. Hall KD, Ayuketah A, Brychta R, Cai H, Cassimatis T, Chen KY, et al. 27. Mallick H, Rahnavard A, McIver LJ, Ma S, Zhang Y, Nguyen Ultra-processed diets cause excess calorie intake and weight gain: an inpatient LH, et al. Multivariable association discovery in population-scale randomized controlled trial of ad libitum food intake. Cell Metab. (2019) 30:67– meta-omics studies. Coelho LP, editor. PLoS Comput Biol. (2021) 77.e3. doi: 10.1016/j.cmet.2019.05.008 17:e1009442. doi: 10.1371/journal.pcbi.1009442 14. Schnabel L, Kesse-Guyot E, Allès B, Touvier M, Srour B, Hercberg 28. Konieczna J, Morey M, Abete I, Bes-Rastrollo M, Ruiz-Canela M, Vioque S, et al. Association between ultraprocessed food consumption and risk of J, et al. Contribution of ultra-processed foods in visceral fat deposition and other Frontiers in Nutrition 10 frontiersin.org Atzeni et al. 10.3389/fnut.2022.976547 adiposity indicators: Prospective analysis nested in the PREDIMED-Plus trial. Clin 40. De Filippis F, Pasolli E, Tett A, Tarallo S, Naccarati A, De Angelis M, Nutr. (2021) 40:4290–300. doi: 10.1016/j.clnu.2021.01.019 et al. Distinct genetic and functional traits of human intestinal prevotella copri strains are associated with different habitual diets. Cell Host Microbe. (2019) 29. Mendonça R de D, Pimenta AM, Gea A, de la Fuente-Arrillaga C, Martinez- 25:444–53.e3. doi: 10.1016/j.chom.2019.01.004 Gonzalez MA, Lopes ACS, et al. Ultraprocessed food consumption and risk of overweight and obesity: the University of Navarra Follow-Up (SUN) cohort study. 41. Mukhopadhya I, Hansen R, Nicholl CE, Alhaidan YA, Thomson JM, Berry Am J Clin Nutr. (2016) 104:1433–40. doi: 10.3945/ajcn.116.135004 SH, et al. A comprehensive evaluation of colonic mucosal isolates of sutterella wadsworthensis from inflammatory bowel disease. Bereswill S, editor. PLoS One. 30. Elizabeth L, Machado P, Zinöcker M, Baker P, Lawrence M. Ultra- (2011) 6:e27076. doi: 10.1371/journal.pone.0027076 processed foods and health outcomes: A narrative review. Nutrients. (2020) 12:1– 36. doi: 10.3390/nu12071955 42. Hiippala K, Kainulainen V, Kalliomäki M, Arkkila P, Satokari R. Mucosal prevalence and interactions with the epithelium indicate commensalism of 31. Miclotte L, Van de Wiele T. Food processing, gut microbiota Sutterella spp. Front Microbiol. (2016) 7:1–13. doi: 10.3389/fmicb.2016.01706 and the globesity problem. Crit Rev Food Sci Nutr. (2020) 60:1769– 82. doi: 10.1080/10408398.2019.1596878 43. Kaakoush NO. Sutterella species, IgA-degrading bacteria in ulcerative colitis. Trends Microbiol. (2020) 28:519–22. doi: 10.1016/j.tim.2020.02.018 32. Cust AE, Skilton MR, van Bakel MME, Halkjær J, Olsen A, Agnoli C, et al. Total dietary carbohydrate, sugar, starch and fibre intakes in the European 44. Derqaoui S, Oukessou M, Attrassi K, Elftouhy FZ, Nassik S. Detection Prospective Investigation into Cancer and Nutrition. Eur J Clin Nutr. (2009) of Sutterella spp. in Broiler Liver and Breast. Front Vet Sci. (2022) 63:S37–60. doi: 10.1038/ejcn.2009.74 9:e859902. doi: 10.3389/fvets.2022.859902 33. De Filippis F, Pellegrini N, Vannini L, Jeffery IB, La Storia A, 45. Hou JK, Abraham B, El-Serag H. Dietary intake and risk of developing Laghi L, et al. High-level adherence to a Mediterranean diet beneficially inflammatory bowel disease: a systematic review of the literature. Am J impacts the gut microbiota and associated metabolome. Gut. (2016) 65:1812– Gastroenterol. (2011) 106:563–73. doi: 10.1038/ajg.2011.44 21. doi: 10.1136/gutjnl-2015-309957 46. Wang F, Feng J, Gao Q, Ma M, Lin X, Liu J, et al. Carbohydrate 34. Partula V, Mondot S, Torres MJ, Kesse-Guyot E, Deschasaux M, Assmann and protein intake and risk of ulcerative colitis: Systematic review and dose- K, et al. Associations between usual diet and gut microbiota composition: results response meta-analysis of epidemiological studies. Clin Nutr. (2017) 36:1259– from the Milieu Intérieur cross-sectional study. Am J Clin Nutr. (2019) 109:1472– 65. doi: 10.1016/j.clnu.2016.10.009 83. doi: 10.1093/ajcn/nqz029 47. Nie J-Y, Zhao Q. Beverage consumption and risk of ulcerative colitis. 35. Tian T, Zhang X, Luo T, Wang D, Sun Y, Dai J. Effects of short-term dietary Medicine. (2017) 96:e9070. doi: 10.1097/MD.0000000000009070 fiber intervention on gut microbiota in young healthy people. Diabetes, Metab 48. Li F, Liu X, Wang W, Zhang D. Consumption of vegetables and fruit Syndr Obes Targets Ther. (2021) 14:3507–16. doi: 10.2147/DMSO.S313385 and the risk of inflammatory bowel disease. Eur J Gastroenterol Hepatol. (2015) 36. Gryaznova MV, Solodskikh SA, Panevina AV, Syromyatnikov MY, 27:623–30. doi: 10.1097/MEG.0000000000000330 Dvoretskaya YD, Sviridova TN, et al. Study of microbiome changes in patients 49. Gomez-Arango LF, Barrett HL, McIntyre HD, Callaway LK, Morrison M, with ulcerative colitis in the Central European part of Russia. Heliyon. (2021) Dekker Nitert M. Increased systolic and diastolic blood pressure is associated with 7:e06432. doi: 10.1016/j.heliyon.2021.e06432 altered gut microbiota composition and butyrate production in early pregnancy. 37. Wang W, Pan Y, Wang L, Zhou H, Song G, Wang Y, et al. Optimal Hypertension. (2016) 68:974–81. doi: 10.1161/HYPERTENSIONAHA.116.07910 dietary ferulic acid for suppressing the obesity-related disorders in leptin- 50. Mendonça R de D, Lopes ACS, Pimenta AM, Gea A, Martinez-Gonzalez deficient obese C57BL/6J -ob/ob mice. J Agric Food Chem. (2019) 67:4250– MA, Bes-Rastrollo M. Ultra-processed food consumption and the incidence of 8. doi: 10.1021/acs.jafc.8b06760 hypertension in a mediterranean cohort: the seguimiento universidad de navarra 38. Dahl WJ, Rivero Mendoza D, Lambert JM. Diet, Nutrients and the project. Am J Hypertens. (2016) 2016:hpw137. doi: 10.1093/ajh/hpw137 Microbiome. Progress in Molecular Biology and Translational Science. Elsevier Inc. 51. Ascher S, Reinhardt C. The gut microbiota: An emerging risk factor (2020). doi: 10.1016/bs.pmbts.2020.04.006 for cardiovascular and cerebrovascular disease. Eur J Immunol. (2018) 48:564– 39. Precup G, Vodnar DC. Gut Prevotella as a possible biomarker of diet and its 75. doi: 10.1002/eji.201646879 eubiotic versus dysbiotic roles: A comprehensive literature review. Br J Nutr. (2019) 52. Shim J-S, Oh K, Kim HC. Dietary assessment methods in epidemiologic 122:131–40. doi: 10.1017/S0007114519000680 studies. Epidemiol Health. (2014) 2014:e2014009. doi: 10.4178/epih/e2014009 Frontiers in Nutrition 11 frontiersin.org