**1. Introduction**

Multiple sclerosis (MS) is a chronic inflammatory, neurodegenerative disease, mediated by autoimmune reactions against myelin proteins and gangliosides in white and grey matter of the brain and spinal cord, promoting physical disability, cognitive impairment, and decreased quality of life in young adults, aged between 20 and 40 years [1,2]. The incidence of MS is increasing worldwide and estimated to range from 5 to 300 per 100,000

**Citation:** Pellizoni, F.P.; Leite, A.Z.; Rodrigues, N.d.C.; Ubaiz, M.J.; Gonzaga, M.I.; Takaoka, N.N.C.; Mariano, V.S.; Omori, W.P.; Pinheiro, D.G.; Matheucci Junior, E.; et al. Detection of Dysbiosis and Increased Intestinal Permeability in Brazilian Patients with Relapsing–Remitting Multiple Sclerosis. *Int. J. Environ. Res. Public Health* **2021**, *18*, 4621. https://doi.org/10.3390/ijerph18094621

Academic Editors: Diana María Cardona Mena and Pablo Roman

Received: 11 March 2021 Accepted: 14 April 2021 Published: 27 April 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

individuals, affecting females three times more and having a significant socioeconomic impact, with financial burden to patients and to developed and developing economies [2,3].

The MS onset is clinically characterized as relapsing–remitting (RRMS), diagnosed in 85 to 90% of patients [1,4]. The relapses are due to blood–brain barrier breakdown and infiltration of T and B cells and myeloid cells into the central nervous system (CNS) parenchyma, which induces acute inflammation, detected as gadolinium-positive lesions in magnetic resonance imaging (MRI) [3]. Permanent neurological lesions and clinical disability evolve to a secondary progressive form, and few patients present a primary progressive course from disease onset [1]. Complex genetic–environmental interactions are hypothesized to be involved in MS development, including human leukocyte antigen (HLA) genes, Epstein–Barr virus infections, tobacco exposure, obesity, vitamin D deficiency, and alterations of the gu<sup>t</sup> microbiota [1,5,6].

In homeostatic or eubiosis conditions, the gu<sup>t</sup> microbiota is dominated by microorganisms that contribute to food digestion and fermentation, nutrient absorption, vitamin synthesis, epithelial cell maturation, gu<sup>t</sup> barrier integrity, development and education of the immune system, protection against pathogens and inflammation, and regulation of host metabolism and CNS physiology [7–10]. Recently, it has become evident that the gu<sup>t</sup> microbiota can affect neurologic processes through bidirectional communications, involving the enteric nervous system, the endocrine/immune systems, the gu<sup>t</sup> microbiota, and their metabolites [10–13]. Neurotransmitters and short-chain fatty acids (SCFAs), derived from microbiota fermentation, can shape immune responses and impact behavior, memory, and neurodegenerative diseases [10,12,14,15]. Thus, alterations in function and diversity of the gu<sup>t</sup> microbiota, known as dysbiosis, are associated with a dysregulation in these gut–brain connections, increased gu<sup>t</sup> and blood–brain barrier permeability and neuroinflammation and can contribute to the development of inflammatory autoimmune diseases, including MS [16–19].

In MS animal models, when experimental autoimmune encephalomyelitis (EAE) was induced in germ-free mice, a decrease in inflammatory interferon-gamma (IFN-γ) and interleukin-(IL)-17A levels in the CNS was detected, as well as an increase in regulatory T cells (Treg) in the gu<sup>t</sup> mucosa [20]. On the other hand, the colonization of EAE mice with segmented filamentous bacteria induced Th17 differentiation in the lamina propria and migration to the CNS, increasing neuroinflammation and disease severity [20,21]. The disease score ameliorated when germ-free EAE mice were colonized by *Bacteroides fragilis* containing polysaccharide A*,* which induces IL-10-secreting Treg cells and suppress the T-helper (Th)-17 subpopulation [22,23]. Moreover, when fecal samples from MS patients were transferred to germ-free mice, genetically susceptible to EAE, the mice developed the disease and significantly produced less IL-10 than mice colonized with feces from healthy subjects [24]. These data sugges<sup>t</sup> that the gu<sup>t</sup> microbiota is linked to disease severity and immune response during MS development [10].

In humans, the gu<sup>t</sup> microbiota from untreated RRMS patients, from different populations (China, Japan, Germany, USA), differs from that of healthy controls, and patients with active disease present decreased microbiota diversity. Intestinal dysbiosis in MS was predominantly characterized by decreased Firmicutes, Clostridia clusters XIVa and IV, *Faecalibacterium*, *Butyricimonas*, *Prevotella*, and *Lactobacillus* species, and increased abundance of *Pseudomonas*, *Mycoplasma*, *Haemophilus*, *Streptococcus*,*Akkermansia muciniphila*, and *Methanobrevibacter smithii* [24–33]. In addition, MS patients with increased peripheral Th17 lymphocytes and higher disease activity presented an increased Firmicutes/Bacteroidetes ratio, *Streptococcus* amounts, and decreased relative abundance of *Prevotella* species [34]. Interestingly, the taxonomic composition during remission showed richness and evenness similar to those of healthy individuals, and even the frequency of relapses seemed to be influenced by the intestinal microbiota [29,35].

There are few studies evaluating the effect of disease-modifying therapies (DMTs), used to treat MS patients on intestinal microbiota composition. Some studies sugges<sup>t</sup> that these therapies are capable of reversing dysbiosis and restore a "healthy" gu<sup>t</sup> microbiota, similar to that of control subjects [19]. Patients on IFN-β or glatiramer acetate treatment showed increased abundance of *Prevotella, Sutterella*, and *Prevotella copri* and decreased *Sarcina* species and gu<sup>t</sup> microbiota richness [29,36,37]. Besides that, evidence from animal models and human studies demonstrated that gu<sup>t</sup> microbes and their metabolites can influence drug bioavailability, pharmacokinetics, clinical response, as well as adverse events, supporting the importance of studies on the interaction of the gu<sup>t</sup> microbiota with DMTs [38,39]. The future goal is to predict therapeutic responses based on microbiome analyses and propose diet interventions and the use of probiotics or other microbiota modulators as adjuvant therapy to enhance the therapeutic efficacy of DMTs [40,41].

On the basis of this background and the fact that there are no studies evaluating the gu<sup>t</sup> microbiota in Brazilian MS patients, the aim of the present study was to evaluate the gu<sup>t</sup> microbiota in RRMS patients receiving DMTs and correlate these data with dietary habits, clinical parameters, cytokines, and zonulin concentrations.

#### **2. Materials and Methods**

#### *2.1. Selection of Relapsing–Remitting MS Patients and Controls*

Relapsing–remitting multiple sclerosis (RRMS) patients, diagnosed according to the Poser and colleagues criteria [42], were selected by the Neurologist from the School of Health Sciences Dr. Paulo Prata, Barretos, Sao Paulo, Brazil. The Ethics Committee on Human Research from the Barretos Educational Foundation approved the present study (Process number 1522.762/2016), and all subjects signed the informed consent in accordance with the Declaration of Helsinki.

A total of 18 RRMS patients, 16 females and 2 males (mean age − standard deviation (SD) = 46.06 − 11.83 years), were included in this study. Eighteen control subjects, ageand-sex-matched, were included as a control group (mean age − SD = 45.50 − 11.03 years). After the consent, all of subjects answered a food frequency questionnaire (FFQ) that was designed by specialized nutritionists. The FFQ included questions concerning dietary habits, such as consumption of vegetables, fruits, carbohydrates, animal-derived proteins, saturated and trans fats, dairy products, and canned products. The options for frequency of consumption in the FFQ was classified as (1) Never consumes; (2) Less than once a month; (3) One to three times a month; (4) Once or twice a week; (5) Three to five times a week; (6) Six to seven times a week. Data were expressed in percentages based on the responses of patients and controls. Thereafter, peripheral blood (8 mL) was collected, and stool samples were requested and delivered within five days.

At enrollment, exclusion criteria for patients and controls included use of antibiotics and laxatives and vaccination in the last 60 days. Chronic diarrhea and gastrointestinal surgeries, such as bariatric, cholecystectomy, and appendectomy, were also considered as exclusion criteria for both groups.

Clinical data from MS patients, such as body mass index (BMI), disease duration, Expanded Disability Status Score (EDSS), presence/absence of gadolinium (Gd)-enhanced brain magnetic resonance imaging (MRI) lesions, and disease-modifying therapies (DMTs) were recorded. The mean body mass index of the MS patients was 26. Three patients reported having systemic arterial hypertension, and two patients reported taking vitamin D All other patients included in this study reported no other comorbidity. Demographic characteristics and clinical data from RRMS patients are summarized in Table 1.

.

#### *2.2. Bacterial DNA Extraction, Real-Time PCR, and 16S Sequencing*

DNA was extracted from 200 mg of stool samples by using QIAamp DNA Stool Mini Kit (QIAGEN, Hilden, Germany), according to the manufacturer's instructions. DNA was quantified by Nanodrop and adjusted to 5 ng/mL. Primers were specific for *Bacteroides*, *Bifidobacterium*, *Lactobacillus*, *Prevotella*, and *Roseburia* species [43]. Reactions were performed by using Power SYBR Green PCR Master Mix (Applied Biosystems, Life Technologies, Carlsbad, CA, USA), 2 uM of forward/reverse primers, and 5 ng of DNA. For relative quantification, DNA copy numbers from target primers were normalized for the copy numbers

of universal primer. The relative abundance was calculated by using the cycle-threshold (Ct) values and was expressed by the relative expression units method (REU) [44], per 200 mg of stool.


**Table 1.** Demographic and clinical data of the relapsing–remitting multiple sclerosis patients.

F: Female; M: Male; BMI: Body Mass Index; EDSS: Expanded Disability Status Score; MRI: Magnetic resonance imaging; Gd+: Presence of gadolinium-enhanced brain lesions; ND: not determined; Gd-: Absence of inflammatory active lesions; DMT: Disease-modifying therapy; IFN-*β*-1b: Interferon-*β*-1b; AZA: Azathioprine; GA: Glatiramer acetate; TER: Teriflunomide; FTY720: Fingolimod; NAT: Natalizumab.

> For bacterial 16S sequencing, DNA was quantified by Quantus fluorometer and adjusted to 5 ng/mL using Tris buffer (10 mM, pH 8.5). V3 and V4 regions of the bacterial 16S [45] were amplified by using bacterial DNA, V3/V4 primers, and the 2X KAPA HiFi HotStart Ready Mix (Kapa Biosystems, MA, USA). PCR purification was performed using AMPure XP Beads Kit (BD Biosciences, San Jose, CA, USA). DNA libraries were constructed according to the Illumina protocols, and sequencing was conducted by an Illumina MiSeq platform system.

#### *2.3. Cytokine Determination by Cytometric Bead Array*

After peripheral blood collection (8 mL) in gel tubes with clot activator, samples were incubated for 50 min and then centrifuged at 1372 g for 5 min, 25 ◦C. Isolated serum samples were stored until cytokine determination. Cytokine detection was performed by using a cytometric bead array (Human Th1/Th2/Th17 Cytokine Kit, BD Biosciences, Franklin Lakes, NJ, USA). Serum levels of IL-2, IL-4, IL-6, IL-10, IL-17A, tumor necrosis factor (TNF), and IFN-γ were determined by flow cytometer FACSCanto™ II (BD Biosciences). Analyses were performed by BDFCAP array™ software, and data were expressed in pg/mL.

#### *2.4. Zonulin Serum Quantification by Sandwich ELISA*

Serum samples were isolated from peripheral blood collected in gel tubes with clot activator. After collection, samples were incubated for 50 min, centrifuged at 1372 g for 5 min, and stored until zonulin determination. A human Zonulin ELISA Kit (Elabscience, MD, USA) was used to quantify zonulin concentrations. Plates were pre-coated with antibodies to human zonulin, and serum samples and standards were incubated for 1 h, 37 ◦C. Then, incubation with biotinylated detection antibodies and avidin–horseradish peroxidase conjugate was performed for 30 min. Three washing steps followed to remove unbound and free molecules. The substrate solution was added to each well and incubated for 15 min. The enzyme–substrate reaction was blocked by a stop solution, and the color turned yellow. The optical density was measured in a spectrophotometer at 450 nm. A standard curve was constructed, and zonulin concentrations were calculated by converting the obtained optical density in ng/mL.
