Received: 14 February 2023 | Revised: 3 May 2023 | Accepted: 20 August 2023

DOI: 10.1002/fsn3.3654

ORIGINAL ARTICLE

Lemon flavonoids nutraceutical (Eriomin®) attenuates

prediabetes intestinal dysbiosis: A double-­blind randomized
controlled trial

Fernanda M. M. Ramos1 | Carolina B. Ribeiro1 | Thais B. Cesar1 | Dragan Milenkovic2 |

Lucélia Cabral3 | Melline F. Noronha4 | Katia Sivieri1

1
Graduate Program in Food, Nutrition
and Food Engineering, Sao Paulo State Abstract
University (UNESP), Araraquara, Brazil
Eriocitrin (eriodictyol 7-­O-­β-­rutinoside), a citrus flavonoid from lemon juice and peel,
2
Department of Nutrition, University of
California Davis, Davis, California, USA
reduces hyperglycemia and improves diabetes-­related biomarkers in prediabetes pa-
3
Institute of Biosciences, Depart of tients. Eriocitrin is first metabolized by gut microbiota, producing energy for gut cells
General and Applied Biology, São Paulo and short chain fatty acids that play a relevant role in glycemic control. The aim of
State University (UNESP), Rio Claro, Brazil
4 this study was to assess the effect of Eriomin®, a nutraceutical composed of 70%
Research Informatics Core, Research
Resource Center, University of Illinois at eriocitrin, 5% hesperidin, and 4% naringin, on the microbiota of prediabetic patients.
Chicago, Chicago, Illinois, USA
Patients were randomly divided into two groups and received unlabeled capsules of
Correspondence Eriomin® (200 mg/day) or placebo during 12 weeks. After treatment with the nutra-
Thais B. Cesar, Graduate Program in Food,
ceutical, it was a 6% decrease of hyperglycemia and 22% increase of GLP-­1 blood
Nutrition and Food Engineering, Sao Paulo
State University (UNESP), Araraquara, levels of (p < .05). The profile of intestinal microorganisms, obtained by 16S rRNA se-
Brazil.
quencing of the patients' feces extract, showed changes in microbiota composition,
Email: thais.cesar@unesp.br
such as lower growth of Firmicutes and less abundance of the Lachnospiraceae fam-
Funding information
ily. The family Ruminococcaceae increased and Blautia genus reduced with Eriomin®
Coordenação de Aperfeiçoamento de
Pessoal de Nível Superior; Ingredients by supplementation. In additional, Blautia was positively correlated with hyperglycemia
Nature
reduction. In conclusion, the nutraceutical Eriomin® moderately reduced the growth
of microorganisms associated with intestinal dysbiosis and increased the abundance
of beneficial bacteria. Changes promoted mainly by the flavonoid eriocitrin in the mi-
crobiota were related to a lower glycemic level and increased production of GLP-­1 in
patients with prediabetes.

KEYWORDS
clinical trial, eriocitrin nutraceutical, Eriomin®, intestinal dysbiosis, lemon flavonoid,
microbiota (16S rRNA), pre-­diabetes

1 | I NTRO D U C TI O N glucose, dyslipidemia, and prehypertension are frequent and mea-

surable signs that precede the development of type 2 diabetes,
Metabolic disorders in adults and the elderly often signal the onset atherosclerosis, and cardiovascular disease (Beulens et al., 2019).
of chronic degenerative diseases, with a great impact on the health Prediabetes can be detected by a slight increase in fasting
and longevity of affected individuals. For example, elevated blood blood glucose of 5.6–­6.9 mmol/L, as well as glucose intolerance

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,
provided the original work is properly cited.
© 2023 The Authors. Food Science & Nutrition published by Wiley Periodicals LLC.

Food Sci Nutr. 2023;11:7283–7295.  wileyonlinelibrary.com/journal/fsn3 | 7283

7284 | RAMOS et al.

(7.8–­11 mmol/L), followed or not by an increase in glycated he- 2 | M ATE R I A L S A N D M E TH O DS

moglobin (HbA1c >5.7–­6.4%) (American Diabetes Association
Professional Practice Committee et al., 2022). In addition, other 2.1 | Participants
metabolic changes, such as excess visceral fat, chronic low-­grade
systemic inflammation, and disruption of healthy gut microbiota, Individuals of both sexes, aged 30–­69 years, identified with predia-
known as dysbiosis, constitute primary causes of degenerative dis- betes, were eligible to participate in the study. They presented at
eases (Lloyd-­Price et al., 2016). least one of the following inclusion criteria: (1) fasting blood glucose
Microbiota dysbiosis reflects changes in the resistance, resil- of 5.6–­6.9 mmol/L, (2) glucose intolerance, as assessed by glucose
ience, stability, and diversity of intestinal microorganisms, as well tolerance test (7.8–­11 mmol/L); (3) glycated hemoglobin (HBA1C)
as quantitative and qualitative reduction in the abundance of short-­ of 5.7–­6.4% (American Diabetes Association Professional Practice
chain fatty acids (SCFAs) produced by gut bacteria (Iebba et al., 2016). Committee et al., 2022). However, individuals under the use of
Associated with this imbalance, the fragility of the intestinal barrier hypoglycemic drugs, weight loss drugs, antibiotics (within the last
associated with fewer tight junctions increases the permeability and 3 months), laxatives, and/or dietary supplements (vitamins, miner-
translocation of lipopolysaccharides (LPS) and Gram-­negative bacte- als, and bioflavonoids) were excluded. In addition, individuals under
rial wall fragments, which reach the bloodstream causing metabolic intense physical training (more than 10 h/week) or with chronic dis-
endotoxemia (Cani, 2018; Zhang et al., 2013). This event stimulates eases, such as type 2 diabetes, liver disease, renal syndrome, cardio-
the secretion of proinflammatory cytokines and activates Toll-­like vascular disease, polycystic ovary syndrome, or thyroid dysfunction,
receptor 4 (TLR-­4) and nuclear factor kappa B (NF-­κB) in epithelial were excluded. Those with gastrointestinal diseases such as malab-
cells, triggering low-­grade systemic inflammation, common in obese sorption syndrome, chronic inflammatory bowel disease, colorectal
animals and humans (Gomes et al., 2017). In turn, inflammation ac- cancer, celiac disease, diverticulitis, and Crohn's disease, which af-
tivates macrophages infiltration in adipose tissue, liver, and kidneys, fect the gut microbiota, were also excluded. Smokers, individuals
stimulating the production of tumor necrosis factor α (TNF-­α), in- with a history of drug or alcoholism, and pregnant women were also
terleukins (such as IL-­6), adipokines (such as leptin and resistin), and unable to participate.
C-­reaction protein (CRP), causing insulin resistance and leading to Sixty-­t wo eligible individuals were selected through medical re-
prediabetes, among other metabolic disorders (Boutagy et al., 2016; cords from the public health system of Araraquara SP Brazil. They
Zhou et al., 2019). were contacted by phone by the study personnel to explain the
Reversal of prediabetic intestinal dysbiosis may be a strategy to objectives of the study, as well as obtain acceptance to voluntarily
combat the development of type 2 diabetes mellitus (Cani, 2018). participate. Forty-­five volunteers met the inclusion/exclusion crite-
One attempt would be to modulate the gut microbiota through sup- ria and were invited to complete a clinical data questionnaire and to
plementation with phenolics, which are found in a wide variety of have blood samples withdrawn for basal tests. Of these, 29 com-
foods such as fruits, vegetables, herbs, seeds, and cereals, as well as pleted the study, 16 women and 13 men, aged 50 ± 10 years.
in beverages such as coffee, tea, cocoa, and wine (Kasprzak-­Drozd The experimental protocol was approved by the Ethics Commit-
et al., 2021). Among them, citrus-­derived flavonoids such as hesper- tee of the Pharmacy School, UNESP (CAEE: 67610217.6.0000.5426)
idin, naringenin, and eriocitrin appear to benefit the health of the and registered at Clini​calTr​ials.gov (NCT03925909). The procedures
microbiota (Amiot et al., 2016; Ávila-­Gálvez et al., 2021). performed followed the ethical guidelines of the National Health
Eriocitrin (eriodictyol 7-­O-­β-­rutinoside), a citrus flavonoid found Council (Res. 466/12) and the Declaration of Helsinki (1964), and
primarily in lemon peel, is recognized for its anti-­inflammatory and all participants voluntarily signed an informed consent form before
antioxidant properties, and it has the potential to be used in diabetes starting the study.
therapy (Miyake et al., 1998). After its ingestion, most of the erioc-
itrin passes intact through the small intestine and is metabolized in
the large intestine by the gut microbiota (Ávila-­Gálvez et al., 2021). 2.2 | Study design
We recently tested the efficacy of the citrus flavonoid eriocitrin, in
reversing hyperglycemia in patients treated with the nutraceutical A double-­blind, randomized, placebo-­controlled study began with
Eriomin® (Ribeiro et al., 2019), composed mainly of eriocitrin (70%), 45 individuals randomly divided into two groups: Eriomin® and
hesperidin, naringin, and didymin (20%), and fiber plant material placebo. Each patient, out of a total of 45, was numbered accord-
(10%). The results showed a significant reduction in blood glucose ing to their casual entry into the trial, from 1 to 45 subject. Then,
associated with an increase in glucagon-­like peptide-­1 (GLP-­1) and following a random list of 45 numbers, the first 25 were chosen to
a reduction in inflammatory biomarkers. Although an improvement be the patients in the experimental group. The remaining 20 ran-
of glycemic profile and a reversion of prediabetes were for 24% of dom numbers were assigned to patients in the placebo group. All
patients with Eriomin®, its impact on the gut microbiota has not yet subjects were treated with a 200 mg/day of Eriomin® (n = 25) or the
been studied. Thus, the aim of this study was to evaluate the potential placebo (n = 20). They received instructions to maintain their usual
of Eriomin® to modulate the intestinal microbiota and its relationship diet during the intervention and minimize the intake of citrus foods
with the improvement of blood glucose in prediabetic individuals. and flavonoid-­rich beverages, such as coffee, tea, cocoa, fruit juice,
RAMOS et al. | 7285

and wine. Eriomin® and placebo capsules in unlabeled vials were naringin, and plus 2 mg of didymin. The placebo was composed of
provided to the participants every 2 weeks. The volunteers took one pure microcrystalline corn starch. Placebo and Eriomin® were en-
capsule per day with water after dinner for 12 weeks. Sixteen sub- capsulated in tablets of the same size, shape, and color, by a reg-
jects were excluded during the experimental period due to moving istered pharmacist, and the appearance was identical between for
away (n = 2), incorrect or irregular use of capsules (n = 10), use of an- both placebo and the active supplement.
tibiotics due to an unrelated problem (n = 1), and intentional weight
loss (n = 3). Twenty-­nine subjects who followed the experimental
protocol recommendations were included in the final analysis. The 2.4 | Sampling and storage of blood and feces
chosen dose of Eriomin® (200 mg/day) was based on our previous
study (Ribeiro et al., 2019), which showed no difference in lowering The 12 h fasting blood samples were collected at the beginning of
blood glucose after 200, 400, or 800 mg of Eriomin® for 12 weeks the first week and at the end of the 12th week of the protocol, and
(Ribeiro et al., 2019). The experimental design is shown on Figure 1. the blood serum samples were kept at −80°C. The collection and
storage were performed at the Clinical Analysis Laboratory São
Lucas, Araraquara-­SP. Fecal samples were self-­collected by the vol-
2.3 | Supplement preparation unteers in a commercial sterile container, provided by the laboratory,
at the beginning of the first week and at the end of the 12th week.
The intervention product was Eriomin®, a supplement of citrus fla- The samples, delivered to the laboratory within 24 h of collection,
vonoids supplied by Ingredients by Nature TM, Montclair, CA. Pu- were homogenized, stored in sterile plastic tubes (10 mL), and kept at
rity, as determined by HPLC, was 70% eriocitrin, 5% hesperetin, 4% −80°C in a deep freezer (Haier Biomedical) until analysis.
naringin, 1% didymin, and 20% of fiber material from epidermis and
periderm cell-­wall, composed by suberin, cutin, lignin, pectin, and
cellulose. The dose of 200 mg of active ingredients per capsule was 2.5 | Microbiota
equal to 140 mg of eriocitrin, plus 10 mg of hesperidin, plus 8 mg of
DNA extraction from stool samples was performed using the
“PowerLyzer @PowerSoil DNA Isolation Kit” (Qiagen). Primers
319F/806R were used to amplify the V3–­V4 region of the 16S
rRNA. In step 1, the forward and reverse primers contained a
sequence of Illumina tags, a variable-­length spacer, a binding se-
quence, and the target sequence 16S to increase diversity and im-
prove the quality of sequencing and execution. Each 25 μL PCR
reaction contained a Kapa2G Robust Hot Start Polymerase unit
(Kapa Biosystems), 1.5 mmol/L MgCl2 , 0.2 mmol/L dNTP mixture,
0.2 μmol/L of each primer, and 1 μL of DNA for each sample. In
step 2, each sample was encoded with a unique back-­and-­forth
barcode combination with an Illumina P5 adapter string, a unique
8 nt barcode, a partial match string of the lead adapter used in step
1, and reverse primers with an Illumina P7 adapter. The PCR reac-
tion in step 2 contained a final concentration of 0.2 μmol/L of each
primer with a unique barcode and 1 μL of the product of the PCR
reaction in step 1. The final product was quantified on the Qubit
instrument using the kit of Qubit Broad Range DNA (Invitrogen),
and the individual amplicons were pooled in equal concentrations.
The pooled library was cleaned using Ampure XP beads (Beckman
Coulter), and the band of interest was subjected to isolation by
1.5% gel electrophoresis (Sage Science). The library was quanti-
fied via qPCR followed by 300 bp paired-­e nd sequencing using an
Illumina MiSeq instrument at the Genome Center DNA Technolo-
F I G U R E 1 Experimental design of a 12-­week, double-­blind, gies Core at UC Davis.
randomized, placebo-­controlled, two-­arm parallel study. Of the Demultiplexing the Raw FASTQ files and adjusting the sequence
62 eligible candidates, 45 subjects were selected and separated
adapter were performed using dbcAmplicons v. 0.8.5 (http://github.
randomly into two groups: Placebo (n = 20) and Eriomin® (n = 25).
For 12 weeks they received a daily dose of the designated com/msett​les/dbcAm​plicons). Non-­
immersed direct and reverse
supplement, and after subtracting the dropouts, 29 subjects readings were imported into QIIME2 version 2017.12 (https://
participated in the final analysis of the study. qiime2.org), and the sequence variants were determined following
7286 | RAMOS et al.

the DADA2 analysis pipeline. To classify the sequences according added salt, sugar, seasonings, and sauces. Once completed, a nu-
to their taxonomic information, the q2-­
feature-­
classifier plugin tritionist reviewed the diary with each patient to add any missing
was used according to the VSEARCH alignment method (Gurevich information.
et al., 2013) with the SILVA v132 database (Rognes et al., 2016) and Mean daily food intake was estimated on the basis of the 3 days
with a sequence identity of 99%. The 16S rRNA gene sequencing recorded. The analysis of energy, macronutrient, and micronutrient
analyses were carried in R (Lahti & Eckermann, 2019; McMurdie & intake was performed using the DietBox® software, according to
Holmes, 2013) Studio, version 3.2.4 (R Core Team, 2018) using the the Brazilian Food Composition Table (NEPA-­UNICAMP, 2011). The
Phyloseq and Microbiome packages. quantification of the total energy expenditure (TEE) was estimated
on the basis of the dietary recommended intakes (DRIs) regarding
the overweight and obese population, considering each individu-
2.6 | Biochemical analysis al's sex, age, height, weight, and level of physical activity. Then, the
average energy consumption was compared with the average TEE
Fasting glucose, oral glucose tolerance test, HbA1c, insulin, total (DRI ref).
cholesterol, HDL-­cholesterol, and triglycerides were measured
using commercial kits (Labtest). The homeostasis model assess-
ment of insulin resistance (HOMA-­IR) was estimated using a cutoff 2.9 | Conformity assessment, adverse effects, and
point ≥2.71 (Matthews et al., 1985). The following metabolic and safety of the supplement
inflammatory markers were analyzed by Luminex Milliplex® (RP3X
Scientific): glucagon-­1-­like peptide (GLP-­1), glucagon, C-­peptide, adi- In the biweekly consultation with a nutritionist, participants were
ponectin, tumor necrosis factor alpha (TNF-­α), interleukin 6 (IL-­6), asked to return unused capsules from the previous period, and
and ultrasensitive C-­reactive protein (CRP-­us). Lipid peroxidation they received a new batch for the next 15 days. At each visit, par-
was evaluated using the TBARS assay (Yagi, 1998), and the total an- ticipants were asked about adverse events or discomfort caused
tioxidant capacity was evaluated using the ABTS+ radical assay (Re by the nutraceutical intervention. Adherence to the treatment
et al., 1999). The following hepatic and renal biomarkers were evalu- was assessed by counting capsules untaken and the absence of
ated using commercial kits (Labtest): aspartate transaminase (AST), undesirable effects. Participants who consumed more than 90%
alanine transaminase (ALT), alkaline phosphatase, gamma-­glutamyl of the capsules and who provided and completed all assessments
transferase (γGT), and creatinine. were considered to have good adherence to treatment and were
included in the statistical analysis. The others were removed from
the final analyses.
2.7 | Biometric measurements

The following anthropometric parameters were evaluated using a 2.10 | Randomization and sample size estimation
high-­frequency tetrapolar bioimpedance equipment (Inbody 720®):
weight (kg), height (m), body mass index (BMI) (kg/m2), lean mass The number of participants for this clinical trial was based on the
(kg), fat mass (kg), body fat (%), and visceral fat area (cm2). Partici- mean change in fasting blood glucose after 12 weeks of treatment
pants were instructed to fast for 2 h before bioimpedance, not drink with the nutraceutical Eriomin® (Cesar et al., 2022). In the present
alcohol or caffeine in the last 24 h, wear light clothes, urinate 20 min study, patients with prediabetes were randomized into two groups in
before the test, remove any type of metallic adornment, and not a 1:1 ratio. Randomization sequence and sample size, with a statisti-
practice physical activity in the last 24 h. Waist circumference was cal significance level of 5% and test power of 80%, were estimated
measured at the midpoint between the lower border of the last using Sigma Stat, v.3.11, Systat Software, USA. The minimum sam-
rib and the top of the iliac crest (World Health Organization, n.d.). ple size was estimated at 10 patients per group; and considering a
Blood pressure was measured using a digital monitor (ReliOn, HEM-­ hypothetical dropout rate of 15%, the protocol started with at least
741 CREL). 12 subjects per intervention group, and at the end had at least 10
patients per group.

2.8 | Food and energy consumption

2.11 | Statistical analysis
Participants were instructed to maintain their usual diet and phys-
ical activity throughout the 12 weeks of the experiment. They The data are presented as the mean ± SD. The normality of each set
were also asked to complete a 3-­day food diary (two weekdays and of data was evaluated using the Kolmogorov–­Smirnov or Shapiro–­
one weekend day) during the first and last week of the study. Par- Wilk test. Two-­way ANOVA on ranks, followed by post hoc analysis
ticipants recorded in detail everything they ate or drank, including (Holm–­Sidak method), was used to detect changes in biochemical,
the culinary preparations (fried, boiled, baked, etc.), as well as the clinical, and dietetic factors.
RAMOS et al. | 7287

Unilateral ANOVA (normal data distribution) and Kruskal–­Wallis expected, for prediabetic subjects, fasting glucose, HOMA-­IR,
(non-­normal data distribution) were used to identify differences be- and HbA1c are above the reference values, but after Eriomin®
tween microbiota parameters at the baseline and at the end experiment. treatment, statistical analysis revealed a 6.5% decrease in fasting
Statistical analysis was performed using SigmaStat 3.0 or BioEstat 5.0 blood glucose after 12 weeks (p ≤ .05), while patients who received
software, and statistical significance considered was p ≤ .05. the placebo remained with unchanged levels over time (Table 2
Pearson correlation analysis was used to evaluate the correla- and Figure 2a). Conversely, blood serum GLP-­1 levels in patients
tion between gut microbiota and biochemical variables (fasting glu- treated with Eriomin® significantly increased by 22% (p ≤ .05),
cose) in the beginning and after 12 weeks of Eriomin® treatment. whereas the placebo intervention revealed no change (p ≥ .05;
A correlation network plot was generated, and correlation magni- Table 2 and Figure 2b). Mean blood levels of total cholesterol and
tudes >1.0 (strong co-­abundance relationships) and <−1.0 (strong co-­ LDL-­cholesterol were borderline high, indicating a high risk of de-
exclusion relationships) were plotted. Visualization of the network veloping atherosclerosis, but with some protection provided by
was performed by XLSTAT software. HDL-­cholesterol above 35 mg/dL (NHLBI, 2013).
Neither Eriomin® nor placebo promoted significant changes in
glucagon, C-­peptide, adiponectin, hsCRP, IL-­6, TNF-­α , antioxidant
3 | R E S U LT S capacity, or lipid peroxidation levels (Table 2). Liver enzymes, ALP,
AST, and ALT, showed higher values in the placebo group from the
Adult men (n = 21) and women (n = 24), aged between 32 and beginning toward the end of the experiment, without any changes in
69 years, selected according to the prediabetes criteria from ADA between. In contrast, γ-­GT increased after 12 weeks in the placebo
(American Diabetes Association Professional Practice Committee group, but not with Eriomin® supplement (Table 2).
et al., 2022), participated in this study. Two registered dietitians fol- Statistical analysis of biometric, hemodynamic, and dietary
lowed them before and during the clinical trial (12 weeks), and 29 parameters of prediabetic volunteers undergoing treatment with
subjects completed all protocol steps and were included in the study. Eriomin® or the placebo for 12 weeks showed no change with the
Table 1 summarizes the baseline parameters of the selected sub- supplements after the treatment period (Table 3).
jects. According to these parameters, on average, patients had fast- The unbalanced and hypercaloric dietary pattern is notice-
ing blood glucose ≥100 mg/dL, HbA1c ≥5.7%, HOMA-­IR ≥2.71, body able by the high energy consumption (>3000 kcal), above the
mass index ≥35 kg/m2, and systolic blood pressure ≥ 120 mmHg. Total Energy Expenditure (TEE = 2400 kcal/day) for these obese
Together, these parameters indicated a prediabetes condition with and sedentary individuals (Table 3). In addition, the distribution
insulin resistance, obesity, prehypertension, and a high risk of devel- of macronutrients in kcal (Acceptable Macronutrient Distribution
oping type 2 diabetes (American Diabetes Association Professional Ranges, AMDR) showed high consumption of total lipids (~40%)
Practice Committee et al., 2022). and saturated fats (~13%), exceeding the recommendation for in-
Blood serum biomarkers before and after Eriomin® and pla- dividuals with cardiovascular risk (25% and 7%, respectively). High
cebo supplementation for 12 weeks are shown in Table 2. As consumption of dietary fats indicated a high-­fat diet, associated

TA B L E 1 Baseline characteristics of prediabetic female and male volunteers before treatment with Eriomin® or Placebo for 12 weeks.

Female Male Total

Prediabetes
Baseline parameters n = 16 n = 13 n = 29 cutoff

Age, years 48.3 ± 10.7 51.9 ± 7.8 49.9 ± 9.6 ≥35

Glucose, mg/dL 100 ± 9 112 ± 9 105 ± 11 ≥100–­125
Insulin, μU/mL 21.6 ± 7.5 15.0 ± 8.5 18.6 ± 8.5 –­
Insulin Resustance (HOMA-­IR) a 5.2 ± 1.5 4.1 ± 2.3 4.7 ± 2.0 ≥2.71
Glycated Hemoglobin (HbA1c), % 5.8 ± 0.3 6.1 ± 0.6 6.0 ± 0.5 ≥5.7%
Triglycerides, mg/dL 154 ± 54 180 ± 91 166 ± 73 ≥250
Total Cholesterol, mg/dL 198 ± 48 198 ± 42 198 ± 45 ≥240
HDL-­cholesterol, mg/dL 46 ± 7 41 ± 8 44 ± 8 <35
Weight, kg 91.5 ± 17.6 96.0 ± 20,6 93.7 ± 18.9 –­
Body Mass Index, kg/m2 34.9 ± 6.4 35.1 ± 6.7 33.4 ± 6.7 ≥25
Waist Circumference, cm 114 ± 24 108 ± 16 111 ± 21 ≥90
Systolic Blood Pressure, mmHg 119 ± 10 122 ± 13 121 ± 12 ≥120
Diastolic Blood Pressure, mmHg 75.5 ± 8.2 79.3 ± 11.6 76.0 ± 10.4 ≥80

Note: Data presented as mean ± SD.

7288 | RAMOS et al.

TA B L E 2 Blood serum biomarkers of prediabetic volunteers treated with Eriomin® or Placebo daily supplements for 12 weeks.

Placebo Eriomin®

Variables 0 week 12°week 0 week 12°week

Glucose, mg/dL 103 ± 8 105 ± 12 106 ± 12 99.0* ± 11.0

Insulin, μU/mL 19.5 ± 8.2 18.4 ± 9.9 18.1 ± 8.9 17.7 ± 8.3
HOMA-­IR 4.97 ± 2.34 4.99 ± 3.47 4.53 ± 1.78 4.37 ± 1.88
HbA1c, % 5.86 ± 0.42 5.79 ± 0.47 6.01 ± 0.53 5.87 ± 0.48
Total Cholesterol, mg/dL 222 ± 55 222 ± 45 185 ± 33 185 ± 39
LDL-­Cholesterol, mg/dL 140 ± 40 141 ± 34 102 ± 31 101 ± 35
HDL-­Cholesterol, mg/dL 44.6 ± 10.2 43.7 ± 11.1 44.5 ± 8.5 43.8 ± 7.0
Triglycerides, mg/dL 167 ± 75 189 ± 75 165 ± 74 168 ± 65
Glucagon-­like peptide-­1 (GLP1) ρg/mL 7.96 ± 2.06 7.97 ± 1.82 8.15 ± 2.3 10.0* ± 2.2
Glucagon, ρg/mL 147 ± 13 142 ± 18 144 ± 17 130 ± 11
Adiponectin, μg /mL 18.5 ± 5.21 19.6 ± 4.8 18.1 ± 9.4 20.4 ± 6.4
C Peptide, ρg/mL 1779 ± 483 1743 ± 532 2290 ± 861 2009 ± 671
hsCRP, mg/dL 0.39 ± 0.25 0.32 ± 0.15 0.37 ± 0.22 0.33 ± 0.18
IL-­6, ρg/mL 7.62 ± 5.34 7.61 ± 6.14 5.86 ± 2.50 4.87 ± 1.74
TNF-­α , ρg/mL 5.45 ± 1.35 5.18 ± 1.75 5.66 ± 1.36 5.03 ± 1.51
Antioxidant capacity, μM 1.88 ± 0.04 1.89 ± 0.03 1.86 ± 0.10 1.90 ± 0.03
Lipid peroxidation (MDA) mM 2.25 ± 1.22 2.05 ± 1.21 1.60 ± 0.76 1.49 ± 0.68
Alkaline phosphatase (ALP) U/L 67.3 ± 12.1 67.3 ± 18.0 57.6 ± 13.2 57.3 ± 14.5
γ-­glutamyl transferase (γGT), U/L 57.2 ± 34.8 68.3* ±37.0 38.8 ± 32.4 39.2 ± 29.0
Aspartate transaminase (AST), U/L 26.6b ± 8.2 29.9b ± 8.0 19.4a ± 4.0 21.1a ± 4.5
b b a
Alanine transaminase (ALT), U/L 35.3 ± 17.4 40.1 ± 14.9 20.2 ± 6.3 22.4a ± 8.3
Urea, mg/dL 31.5 ± 8.7 32.5 ± 9.4 34.2 ± 15.0 32.3 ± 9.0
Creatinine, mg/dL 0.90 ± 0.15 0.92 ± 0.23 0.85 ± 0.29 0.86 ± 0.21

Note: Data are presented as mean ± SD. Statistical analysis: Two-­way ANOVA on Ranks, followed by Holm–­Sidak to detect changes in intragroup
(week 0 vs. week 12) and intergroups (Placebo vs. Eriomin).
*Statistical difference significant intragroup (week 0 vs. week 12) (p ≤ .05).
a≠b
Statistical difference significant intergroups (Eriomin vs. Placebo) (p ≤ .05).

with high consumption of cholesterol (>500 mg/day). In addition, supplementation improved the intestinal microbiota, but no change
there was a low consumption of dietary fiber (17 g/day), whose was observed in the Placebo group.
recommendation by American Heart Association (AHA) for adults Microorganism phyla distribution of prediabetic patients shown
is ≥30 g/day (Cesar et al., 2022). small differences between Eriomin® and placebo throughout the ex-
perimental period (Figure 3). In general, both groups showed greater
abundance of the phylum Firmicutes (≥90%), before and after supple-
3.1 | Effects of Eriomin® on gut microbiome mentation. Phylum Actinobacteria was the second most abundant, in
average 3.5% of the total phyla, and after 12 weeks increased 1.5%
A total of 5,187,450 high-­quality readings were obtained from mi- with Eriomin®, but decreased 0.6% with placebo. Together, Bacte-
crobiota samples collected at the beginning and end of the placebo roidetes and Proteobacteria accounted for less than 3% of the total
and Eriomin® treatments. After normalizing the data, 868,564,000 phyla abundance, and decreased slightly in both groups. This profile
sequences were produced. Alpha diversity showed no significant shows a microbiota in dysbiosis (Figure 3).
changes in OTU richness, Chao1, Simpson, and Shannon indices be- The taxonomic assignment performed at the family and genus level
tween the Eriomin® and placebo groups (Table 4). of the prediabetes microbiota showed similar distribution between
Beta diversity based on PERMANOVA statistical analysis re- placebo and Eriomin® (Figures 4a and 5a), with higher abundance of
vealed significant differences in bacterial communities between Lachnospiraceae family (>50%) and Blautia genus (>35%) before sup-
0 and 12 weeks of Eriomin® supplementation (p = .039), whereas plementation. Ruminococcaceae family increased 38% after 12 weeks
the placebo group showed no statistical difference (p = .998; of Eriomin®, in contrast with a decrease of 31% with placebo (p = .04;
Table 5). These results indicate that the 12 weeks of Eriomin® Figure 4b). At the genus level with Eriomin® treatment, it was a
RAMOS et al. | 7289

Our previous clinical study (Ribeiro et al., 2019) showed the ef-
fectiveness of Eriomin® in reducing fasting glucose, glucose intoler-
ance, insulin resistance (HOMA-­IR), glycated hemoglobin (HbA1c),
glucagon, C-­peptide, HSCRP, IL-­6, and TNF-­α . Increasing doses of
Eriomin® were tested (200, 400, and 800 mg/day), but the lowest
dose of 200 mg/day was sufficient to achieve all critical benefits,
such as an increase in GLP-­1, an improvement in the inflammatory
condition and metabolic rate, and a reversal of hyperglycemia in
24% of treated patients (Ribeiro et al., 2019). Likewise, an exper-
imental study carried out in our laboratory with rodents (Ferreira
et al., 2020) showed that a lower dose of Eriocitrin (25 mg/kg) was
more effective in attenuating the metabolic impairment of glucose
and lipids in mice treated with an obesogenic diet than a higher dose
of 100 mg/kg. Currently, the prediabetic and diabetic state has been
characterized by presenting microbiota in dysbiosis, which translates
into a disturbance in the abundance and variety of microorganisms in
the large intestine. However, recent studies have shown that citrus-­
derived flavonoids such as hesperidin, naringenin, and eriocitrin help
balance the microbiota species (Ávila-­Gálvez et al., 2021). Thus, on
the basis of the metabolic benefits obtained previously in predia-
betic and diabetic patients treated with Eriomin® (Cesar et al., 2022;
Ribeiro et al., 2019), we aimed to verify the effect of the nutraceu-
tical Eriomin® on the intestinal dysbiosis, typical of this condition.
At the beginning of this clinical trial, all prediabetic patients had
F I G U R E 2 Course of (a) blood serum glucose (mg/dL) and (b) intestinal microbiota with predominance of the Firmicutes phylum
GLP-­1 over 12 weeks of treatment (Week 0 and Week 12) with
to Bacteroidetes with a ratio of 63:1 and 77:1, respectively, for the
200 mg/day of Eriomin versus Placebo in prediabetes patients.
Circular points represent individual measurements of Glucose (a) Eriomin and placebo groups. The prominence of Firmicutes to Bacte-
and GLP-­1 (b). Dash line represents the average trend line over time. roidetes typically demonstrates a microbiota in dysbiosis, common in
diabetics and prediabetics, as well as in individuals with excess body
weight and body fat (Zhang et al., 2013). After 12 weeks of treat-
decreased of Blautia in contrast with placebo (p = .002), and lower lev- ment, the ratio of Firmicutes to Bacteroidetes increased by 25% with
els of Prevotella genus, while placebo it not changed (p = .04; Figure 5b). Eriomin® and 60% with the placebo, showing that Eriomin® moder-
The correlation analysis between blood glucose and the main genera ated the growth of Firmicutes. The other phyla, Actinobacteria and
of bacteria modulated by Eriomin® identified an interesting correla- Proteobacteria, were not modified by the treatments, and were sta-
tion of the decrease in Blautia and blood glucose after 12 weeks of ble at lower levels across the experiment period. The greater abun-
treatment with Eriomin®, as shown in Figure 6. Therefore, to a limited dance of the phylum Firmicutes, with respect to the other phyla, is
extension, it was seen that Eriomin® modulated one family and two associated with a diet rich in carbohydrates and a high metabolic
bacteria genera of the prediabetes microbiota. efficiency in the energy production of these strains of bacteria.
Concomitantly, the high intake of energy-­rich nutrients by the host,
mainly lipids, was a relevant factor in the maintenance of obesity and
4 | DISCUSSION the presence of inflammatory markers, even after treatment with
Eriomin®. About 25% of the energy consumed in excess was associ-
Short-­term intervention of prediabetic subjects with the nutraceuti- ated with total lipids and saturated fat in the diet. Typically, a hyper-
cal Eriomin® not only reduced hyperglycemia and increased blood lipidemic and hypercaloric diet increases the inflammatory process,
GLP-­1 secretion but also had a significant effect on beta diversity alters the expression of intestinal proteins, and weakens the selec-
in the intestinal colon, changes that were related to the improve- tive permeability of the intestinal barrier. These events in turn lead
ment of prediabetic dysbiosis. This is the first time that the effect of to the translocation of cell-­wall fragments of Gram-­negative bacte-
the citrus flavonoid, eriocitrin, taken as the nutraceutical Eriomin®, ria and lipopolysaccharides, which fuel low-­grade inflammation with
has been studied on the human microbiota to treat hyperglycemia a direct impact on blood glucose dysregulation (Ferreira et al., 2021).
in prediabetic adults. Previous clinical studies with eriocitrin (Cesar Simultaneously, an abundance of 60% Lachnospiraceae, a repre-
et al., 2022; Ribeiro et al., 2019) and data from rodents (Ferreira sentative family of the phylum Firmicutes, was detected in both groups
et al., 2020; Miyake et al., 1998) showed an improvement in glucose at the beginning of the study. This family is associated with glycemic
metabolism with different doses of this compound. dysregulation, metabolic syndrome, and diet-­related obesity, as well
7290 | RAMOS et al.

TA B L E 3 Clinical and dietetic parameters of prediabetic volunteers submitted to Eriomin® or Placebo daily supplements for 12 weeks.

Placebo Eriomin®

Variables 0 week 12°week 0 week 12°week

Biometric
Systolic BP, mmHg 125 ± 10 124 ± 11 102 ± 25 102 ± 26
Diastolic BP, mmHg 78.0 ± 7.9 74.0 ± 7.7 79.0 ± 9.4 77.0 ± 9.5
Body weight, kg 103 ± 17 104 ± 17 102 ± 25 102 ± 26
2
BMI, kg/m 34.5 ± 6.5 34.5 ± 6.7 34.5 ± 7.1 34.5 ± 7.3
Lean mass, kg 34.5 ± 7.3 34.7 ± 7.4 34.9 ± 6.7 34.9 ± 7.3
Fat mass, kg 37.7 ± 8.9 38.3 ± 15.6 40.1 ± 17.8 39.3 ± 18.6
Waist/Hip 1.05 ± 0.10 1.05 ± 0.09 1.06 ± 0.08 1.06 ± 0.09
Nutritional
Energy, kcal/day 3538 ± 932 3500 ± 927 3180 ± 855 3249 ± 1006
(TEE, kcal/day) (2453 ± 320) (2383 ± 392) (2318 ± 356) (2419 ± 306)
Carbohydrates, g/day 380 ± 82 415 ± 126 374 ± 119 360 ± 98
(AMDR = 45–­65%TEE) (44 ± 8%) (48 ± 10%) (47 ± 6%) (46 ± 9%)
Protein, g/day 143 ± 52 138 ± 46 135 ± 45 135 ± 45
(AMDR = 10–­35%TEE) (16 ± 3%) (16 ± 4%) (17 ± 3%) (17 ± 5%)
Lipids 162 ± 66 145 ± 38 129 ± 35 142 ± 76
(AMDR = 20–­35% TEE) (40 ± 8%) (38 ± 11%) (36 ± 6%) (42 ± 25%)
Saturated Fat Acids 37 ± 16 49 ± 12 40 ± 10 44 ± 20
(AHA ≤7% TEE) (13 ± 3%) (13 ± 4%) (11 ± 2%) (13 ± 7%)
Fibers, g/day 17.8 ± 3.3 19.5 ± 3.9 19.6 ± 3.0 19.4 ± 3.8
(AI = 25 g/day) Below Below Below Below
Cholesterol, mg/day 566 ± 188 554 ± 117 553 ± 156 504 ± 118
(AHA ≤ 200 mg/day) Above Above Above Above

Note: Data presented as mean ± SD. The values in (italic) represent % of macronutrient energy (kcal/day) regarding the Total Energy Expenditure
(TEE). No statistic differences detected intragroup (week 0 vs. 12) or intergroups (Eriomin vs. Placebo) (p ≥ .05).
Abbreviations: AHA, American Heart Association; AI, Adequate Intake; AMDR, acceptable macronutrient distribution ranges; TEE, total energy
expenditure (kcal/day).

TA B L E 4 Microbiota alpha diversity

Placebo Eriomin®
indexes (Chao1, Shannon, and Simpson)
0 week 12°week 0 week 12°week p-­Value obtained for all fecal samples before
and after Placebo and Eriomin daily
Chao1 1919 ± 225 1713 ± 232 1745 ± 29 1702 ± 222 .416 supplements.
Shannon 5.9 ± 0.6 5.7 ± 0.6 5.7 ± 0.1 5.7 ± 0.3 .587
Simpson 0.9 ± 0.05 0.9 ± 0.03 0.9 ± 0.004 0.9 ± 0.001 .546
a
p-­values calculated by Kruskal-­Wallis test

TA B L E 5 PERMANOVA statistical analysis from beta-­diversity data using Adonis script in QIIME on week 0 and 12 of treatment (ap ≤ .05).

Groups opts$category Df SumOfSqs MeanSqs F model R2Pr(>F) Pr(>F)

Eriomin® qiime.data$map 5 1.248 0.250 1.118 0.095 0.039a

Residuals 53 11.829 0.223 0.904
Total 58 13.076 1.000
Placebo qiime.data$map 1 0.147 0.147 0.690 0.037 0.998
Residuals 18 3.826 0.213 0.963
Total 19 3.973 1.000
a
p ≤ .05.
RAMOS et al. | 7291

as the prediabetic state (Lippert et al., 2017). However, Eriomin®

treatment slowed the growth of Lachnospiraceae (2%) while placebo
increased it by 11%, showing that Eriomin® attenuated the growth of
this bacteria family. In contrast, the Ruminococcaceae family increased
by 38% with Eriomin®, but decreased by 31% with the placebo. This
family has been associated with the fermentation of complex car-
bohydrates, such as fibers and resistant starch, and polyphenols, to
produce SCFA. Some SCFA, such as propionic acid, can strength the
intestinal barrier and reduce systemic low-­grade inflammation asso-
ciated with glucose intolerance and diabetes (Salonen et al., 2014). A
recent preclinical study that evaluated the metabolism and microbi-
ota modulation of eriocitrin showed that homoeriodictyol, one of the
main metabolites of eriocitrin, was positively correlated with Rumino-
F I G U R E 3 Distribution of intestinal bacterial phyla: Firmicutes, coccus (member of Ruminococcaceae family; Meng et al., 2023).
Actinobacteria, Bacteroidetes, and Proteobacteria before (week 0)
We also observed that the genus Blautia from the Lachnospira-
and after (12 weeks) treatment with Eriomin® or placebo. Data are
expressed as percentage of relative abundance by group mean in ceae family did not change with Eriomin® supplementation, but in-
each experimental period. creased 62% under the placebo, showing a significative difference

F I G U R E 4 (a) Distribution of the

main microbial families before (week
0) and after (week 12) treatment with
Eriomin® or Placebo. Data are expressed
as percentage of relative abundance by
group mean (Eriomin® or Placebo) in
each experimental period. (b) Changes in
the abundance of the Ruminococcaceae
family after 12 weeks of Eriomin®
treatment. Values are expressed in OTU.
The increment of Ruminococcaceae
with Eriomin® versus the decrease
after Placebo is statistically significant
(p ≤ .035).
7292 | RAMOS et al.

F I G U R E 5 (a) Distribution of the

main microbial genera before (week
0) and after (week 12) treatment with
Eriomin® or Placebo. Data are expressed
as percentage of relative abundance by
group mean in each experimental period
(week 0 and week 12). (b) Changes in
the abundance of Blautia genus after
12 weeks of Eriomin® treatment. Values
are expressed in OTU. The increment of
Blautia with Eriomin® versus the decrease
after Placebo is statistically significant
(p ≤ .014).

F I G U R E 6 Heatmap of Pearson's
correlation analysis between gut
microbiota and glucose, after 0 and
12 weeks of Eriomin® treatment.
RAMOS et al. | 7293

between treatments. Blautia is a commensal microorganism and in the glycemic metabolism, showing a potential application in predi-
plays a relevant role for the biotransformation of flavonoids and abetic patients. But, due to the variability of the predominant micro-
SCFA production, helping to maintain intestinal balance (Kim organisms in prediabetic and diabetic microbiota, future studies are
et al., 2014; Liu et al., 2021). It has been suggested that Blautia needed to confirm the effects observed in the current study.
can regulate T cells and decrease intestinal inflammation, and has
potential probiotic properties (Kim et al., 2014). Despite these rel- AUTHOR CONTRIBUTIONS
evant activities in maintaining intestinal health, the relationship of Fernanda M. M. Ramos: Data curation (equal); formal analysis
Blautia with metabolic diseases is still controversial (Liu et al., 2021). (equal); writing –­original draft (equal). Carolina B. Ribeiro: Data
For example, Blautia is abundant in obese and prediabetic patients curation (equal); formal analysis (equal). Thais B. Cesar: Conceptu-
and contains a large amount of potentially pathogenic bacteria alization (equal); formal analysis (equal); funding acquisition (equal);
(Ottosson et al., 2018). Furthermore, Blautia is associated with project administration (equal); writing –­original draft (equal); writ-
high intestinal permeability, which can lead to prediabetes-­related ing –­review and editing (equal). Dragan Milenkovic: Formal analysis
metabolic endotoxemia (Lambeth et al., 2015). Blautia depletion, (equal); writing –­review and editing (equal). Lucelia Cabral: Formal
however, has been described in obese children with metabolic in- analysis (equal). Melline F. Noronha: Formal analysis (equal). Katia
flammation and insulin resistance (Benítez-­Páez et al., 2020). Thus, Sivieri: Conceptualization (equal); methodology (equal); supervision
further studies are needed to identify the species, or strains of (equal); writing –­review and editing (equal).
bacteria, associated with human pathologies. As Eriomin® showed
relative stability in the abundance of Blautia in relation to the pla- AC K N OW L E D G M E N T S
cebo group, it is suggested that the dominant species of Blautia in The authors are very grateful to the volunteers of this study, to the
the prediabetic microbiota would be correlated with the increase São Lucas Clinical Analysis Laboratory (Araraquara, Brazil) for tech-
in glycemia, while the arrest, or decrease, in its growth benefits in- nical assistance, and Veronica Cook for the English review.
sulin sensitivity and the reduction of plasma glucose, as observed
in the positive correlation between the drop in glycemia and the F U N D I N G I N FO R M AT I O N
non-­growth of Blautia in these patients. Furthermore, an excessive This study was funded by a grant from Ingredients by Nature
energy intake (>3000 kcal/day) and high consumption of fats and (Montclair, CA, USA) and supported by a Coordination for the Im-
saturated fatty acids, combined with low intake of dietary fiber provement of Education Personnel (CAPES) scholarship to FR and CR.
(<25 g/day) was observed for all patients. Dietary fiber intake is
recognized as one of the important mechanisms for the health of C O N FL I C T O F I N T E R E S T S TAT E M E N T
the microbiota, as it provides essential nutrients for the growth of The authors declare no conflicts of interest. The sponsors had no
fermentable bacteria, such as Ruminococcaceae family, which are role in the design, execution, interpretation, or writing of the study.
responsible for the production of short-­chain fatty acids (Salonen
et al., 2014). DATA AVA I L A B I L I T Y S TAT E M E N T
The strengths of this study were: the double-­blind and random- Data available on request due to privacy/ethical restrictions.
ized experimental design, with placebo as control, in addition to the
longitudinal monitoring of biochemical and metabolic parameters, E T H I C S S TAT E M E N T
and the bioinformatics of the intestinal microbiota over time as well. The experimental protocol was approved by the Ethics Committee
Among the weaknesses of this study should be mentioned the loss of the Pharmacy School, UNESP (CAEE: 67610217.6.0000.5426)
of participants at the end of the experiment, in addition to the self-­ and registered at Clini​calTr​ials.gov (NCT03215043). The procedures
reported food intake during the experiment. performed followed the ethical guidelines of the National Health
In summary, the nutraceutical Eriocitrin showed beneficial effects Council (Res. 466/12) and the Declaration of Helsinki (1964), and
on the microbiota of prediabetic patients, such as increasing bacte- all participants voluntarily signed an informed consent form before
rial diversity and reducing the growth rate of Firmicutes and Lachno- starting the study.
spiraceae, which are associated with glycemic dysregulation. Blautia
genus, which is linked to inflammatory disorders and altered intesti- ORCID
nal permeability, were also reduced with the nutraceutical treatment. Thais B. Cesar https://orcid.org/0000-0001-7878-7075
Furthermore, Eriomin® increased the abundance of the Ruminococ-
caceae family, which is associated with the production of SCFA and REFERENCES
anti-­inflammatory cytokines. Most of these effects were distinct from American Diabetes Association Professional Practice Committee,
placebo and associated with improvement of glucose metabolism and Draznin, B., Aroda, V. R., Bakris, G., Benson, G., Brown, F. M.,
Freeman, R., Green, J., Huang, E., Isaacs, D., Kahan, S., Leon, J.,
increased levels of GLP1, which has shown to be modulated by specific
Lyons, S. K., Peters, A. L., Prahalad, P., Reusch, J. E. B., Young-­
strains of the gut microbiota (Tomaro-­Duchesneau et al., 2020). These Hyman, D., Das, S., & Kosiborod, M. (2022). Classification and di-
results suggest that lemon flavonoid (Eriomin®) showed a sum of mild agnosis of diabetes: Standards of medical care in diabetes-­2022.
effects on the microbiota, associated with a significant improvement Diabetes Care, 45, S17–­S38. https://doi.org/10.2337/dc22-­S002
7294 | RAMOS et al.

Amiot, M. J., Riva, C., & Vinet, A. (2016). Effects of dietary polyphenols Liu, X., Mao, B., Gu, J., Wu, J., Cui, S., Wang, G., Zhao, J., Zhang, H., &
on metabolic syndrome features in humans: A systematic review. Chen, W. (2021). Blautia-­ a new functional genus with poten-
Obesity Reviews, 17, 573–­586. tial probiotic properties? Gut Microbes, 13(1), 1–­21. https://doi.
Ávila-­Gálvez, M. Á., Giménez-­Bastida, J. A., González-­Sarrías, A., & Espín, org/10.1080/19490​976.2021.1875796
J. C. (2021). New insights into the metabolism of the flavanones Lloyd-­Price, J., Abu-­Ali, G., & Huttenhower, C. (2016). The healthy human
Eriocitrin and hesperidin: A comparative human pharmacokinetic microbiome. Genome Medicine, 8(1), 51. https://doi.org/10.1186/
study. Antioxidants (Basel), 10(3), 435. https://doi.org/10.3390/ s1307​3-­016-­0307-­y
antio​x1003​0 435 Matthews, D., Hosker, J. P., Rudenski, A. S., Naylor, B. A., Treacher, D. F.,
Benítez-­Páez, A., Gómez Del Pugar, E. M., López-­Almela, I., Moya-­Pérez, & Turner, R. C. (1985). Homeostasis model assessment: Insulin re-
Á., Codoñer-­Franch, P., & Sanz, Y. (2020). Depletion of Blautia sistance and β-­cell function from man. Diabetologia, 28(7), 412–­419.
species in the microbiota of obese children relates to intestinal in- https://doi.org/10.1007/BF002​8 0883
flammation and metabolic phenotype worsening. mSystems, 5(2), McMurdie, P. J., & Holmes, S. (2013). Phyloseq: An R package for re-
e00857. https://doi.org/10.1128/mSyst​ems.00857​-­19 producible interactive analysis and graphics of microbiome cen-
Beulens, J., Rutters, F., Rydén, L., Schnell, O., Mellbin, L., Hart, H. sus data. PLoS One, 8(4), e61217. https://doi.org/10.1371/journ​
E., & Vos, R. C. (2019). Risk and management of pre-­diabetes. al.pone.0061217
European Journal of Preventive Cardiology, 26, 47–­54. https://doi. Meng, X., Wu, H., Xiong, J., Li, Y., Chen, L., Gu, Q., & Li, P. (2023).
org/10.1177/20474​87319​880041 Metabolism of eriocitrin in the gut and its regulation on gut mi-
Boutagy, N. E., McMillan, R. P., Frisard, M. I., & Hulver, M. W. (2016). crobiota in mice. Frontiers in Microbiology, 12, 13. https://doi.
Metabolic endotoxemia with obesity: Is it real and is it relevant? org/10.3389/fmicb.2022.1111200
Biochimie, 124, 11–­20. https://doi.org/10.1016/j.biochi.2015.06.020 Miyake, Y., Yamamoto, K., Tsujihara, N., & Osawa, T. (1998). Protective ef-
Cani, P. D. (2018). Human gut microbiome: Hopes, threats and prom- fects of lemon flavonoids on oxidative stress in diabetic rats. Lipids,
ises. Gut, 67(9), 1716–­1725. https://doi.org/10.1136/gutjn​ 33(7), 689–­695. https://doi.org/10.1007/s1174​5-­998-­0258-­y
l-­2018-­316723 NEPA-­UNICAMP. (2011). Tabela Brasileira de Composição de Alimentos/
Cesar, T. B., Ramos, F. M. M., & Ribeiro, C. B. (2022). Nutraceutical NEPA –­ UNICAMP. 4 ed. rev. e ampl. Campinas: NEPA-­UNICAMP,
Eriocitrin (Eriomin®) reduces hyperglycemia by increasing 161p.
glucagon-­like peptide 1 and downregulates systemic inflammation: NHLBI. (2013). ATP III Guidelines At-­A-­Glance Quick Desk Reference,
A crossover-­randomized clinical trial. Journal of Medicinal Food, 25, National Cholesterol Education Program, ATP III At-­A-­Glance:
1–­9. https://doi.org/10.1089/jmf.2021.0181 Quick Desk Reference | NHLBI, NIH (5/24/2022).
Ferreira, P. S., Manthey, J. A., Nery, M. S., & Cesar, T. B. (2021). Ottosson, F., Brunkwall, L., Ericson, U., Nilsson, P. M., Almgren, P.,
Pharmacokinetics and biodistribution of Eriocitrin in rats. Journal Fernandez, C., Melander, O., & Orho-­M elander, M. (2018).
of Agricultural and Food Chemistry, 69(6), 1796–­1805. https://doi. Connection between BMI-­ related plasma metabolite pro-
org/10.1021/acs.jafc.0c04553 file and gut microbiota. The Journal of Clinical Endocrinology
Ferreira, P. S., Manthey, J. A., Nery, M. S., Spolidorio, L. C., & Cesar, and Metabolism, 103(4), 1491–­1501. https://doi.org/10.1210/
T. B. (2020). Low doses of eriocitrin attenuate metabolic impair- jc.2017-­02114
ment of glucose and lipids in ongoing obesogenic diet in mice. R Core Team. (2018). R: A language and environment for statistical comput-
Journal of Nutritional Science, 9, e59. https://doi.org/10.1017/ ing. R Foundation for Statistical Computing. https://www.R-proje​
jns.2020.52 ct.org/
Gomes, J. M. G., Costa, J. A., & Alfenas, R. C. G. (2017). Metabolic endo- Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M., & Rice-­
toxemia and diabetes mellitus: A systematic review. Metabolism, 68, Evans, C. (1999). Antioxidant activity applying an improved
133–­144. https://doi.org/10.1016/j.metab​ol.2016.12.009 ABTS radical cation decolorization assay. Free Radical Biology &
Gurevich, A., Saveliev, V., Vyahhi, N., & Tesler, G. (2013). QUAST: Medicine, 26(9–­10), 1231–­1237. https://doi.org/10.1016/s0891​
Quality assessment tool for genome assemblies. Bioinformatics, -­5849(98)00315​-­3
29(8), 1072–­1075. https://doi.org/10.1093/bioin​forma​tics/btt086 Ribeiro, C. B., Ramos, F. M., Manthey, J. A., & Cesar, T. B. (2019).
Iebba, V., Totino, V., Gagliardi, A., Santangelo, F., Cacciotti, F., Trancassini, Effectiveness of Eriomin® in managing hyperglycemia and rever-
M., Mancini, C., Cicerone, C., Corazziari, E., Pantanella, F., & sal of pre-­diabetes condition: A double blind, randomized, con-
Schippa, S. (2016). Eubiosis and dysbiosis: The two sides of the mi- trolled study. Phytotherapy Research, 33(7), 1921–­1933. https://doi.
crobiota. The New Microbiologica, 39(1), 1–­12. org/10.1002/ptr.6386
Kasprzak-­Drozd, K., Oniszczuk, T., Stasiak, M., & Oniszczuk, A. (2021). Rognes, T., Flouri, T., Nichols, B., Quince, C., & Mahé, F. (2016). VSEARCH:
Beneficial effects of phenolic compounds on gut microbiota and A versatile open-­source tool for metagenomics. PeerJ, 4, e2584.
metabolic syndrome. International Journal of Molecular Sciences, https://doi.org/10.7717/peerj.2584
22(7), 3715. https://doi.org/10.3390/ijms2​2073715 Salonen, A., Lahti, L., Salojärvi, J., Holtrop, G., Korpela, K., Duncan, S. H.,
Kim, C. H., Park, J., & Kim, M. (2014). Gut microbiota-­derived short-­chain & Date, P. (2014). Impact of diet and individual variation on intes-
fatty acids, T cells, and inflammation. Immune Network, 14(6), 277–­ tinal microbiota composition and fermentation products in obese
288. https://doi.org/10.4110/in.2014.14.6.277 men. The ISME Journal, 8(11), 2218–­2230. https://doi.org/10.1038/
Lahti, L., & Eckermann, H. (2019). Introduction to microbiome data sci- ismej.2014.63
ence. Radboud University, Nijmegen. Sep 12–­13, 2019, organized in Tomaro-­Duchesneau, C., LeValley, S. L., Roeth, D., Sun, L., Horrigan, F.
collaboration with University of Turku, Finland. T., Kalkum, M., Hyser, J. M., & Britton, R. A. (2020). Discovery of a
Lambeth, S. M., Carson, T., Lowe, J., Ramaraj, T., Leff, J. W., & Luo, L. bacterial peptide as a modulator of GLP-­1 and metabolic disease.
(2015). Composition, diversity and abundance of gut microbiome Scientific Reports, 10, 4922. https://doi.org/10.1038/s4159​8-­020-­
in pre-­diabetes and type 2 diabetes. Journal of Diabetes and Obesity, 61112​- ­0
2(3), 1–­7. https://doi.org/10.15436/​2376-­0949.15.031 World Health Organization Section 5: Collecting step 2 data: Physical
Lippert, K., Kedenko, L., Antonielli, L., Kedenko, I., Gemeier, C., & Leitner, measurements. http://www.who.int/chp/steps/Part3_Section5.
M. (2017). Gut microbiota dysbiosis associated with glucose me- pdf?ua=1
tabolism disorders and the metabolic syndrome in older adults. Yagi, K. (1998). Simple assay for the level of total lipid peroxides in serum
Beneficial Microbes, 8(4), 545–­556. https://doi.org/10.3920/ or plasma. Methods in Molecular Biology, 108, 101–­106. https://doi.
BM2016.0184 org/10.1385/0-­89603​- ­472-­0:101
RAMOS et al. | 7295

Zhang, X., Shen, D., Fang, Z., Jie, Z., Qiu, X., Zhang, C., Chen, Y., & Ji,
L. (2013). Human gut microbiota changes reveal the progres- How to cite this article: Ramos, F. M. M., Ribeiro, C. B.,
sion of glucose intolerance. PLoS One, 8(8), e71108. https://doi.
Cesar, T. B., Milenkovic, D., Cabral, L., Noronha, M. F., &
org/10.1371/journ​al.pone.0071108
Zhou, W., Sailani, M. R., Contrepois, K., Zhou, Y., Ahadi, S., Leopold, S. Sivieri, K. (2023). Lemon flavonoids nutraceutical (Eriomin®)
R., Zhang, M. J., Rao, V., Avina, M., Mishra, T., Johnson, J., Lee-­ attenuates prediabetes intestinal dysbiosis: A double-­blind
McMullen, B., Chen, S., Metwally, A. A., Tran, T. D. B., Nguyen, H., randomized controlled trial. Food Science & Nutrition, 11,
Zhou, X., Albright, B., Hong, B. Y., … Snyder, M. (2019). Longitudinal
7283–7295. https://doi.org/10.1002/fsn3.3654
multi-­omics of host-­ microbe dynamics in pre-­ diabetes. Nature,
569(7758), 663–­671. https://doi.org/10.1038/s4158​6-­019-­1236-­x