EPILEPSY AND MICROBIOTA: TIME- AND DRUG-RELATED SHIFTS IN THE BACTERIAL COMMUNITY

Epilepsy is a well-known condition characterized by seizures, episodes of electric instability provoking involuntary vigorous shaking in patients. Nowadays there are many strategies, of which the pharmaceutical is the main one, but around one-third of the patients show drug resistance and, despite the different therapeutic approaches for drug-resistant patients, many cannot reach seizure control. Indeed, it is important to look for new complementary therapeutic strategies that can influence the clinical picture and improve the patient’s quality of life. The connection between human microbiota and neurological diseases has been thoroughly studied over the last years. In particular, the gut-brain axis has been addressed to be a key player in the development of the disease and to seizure susceptibility. Nonetheless, there is limited information so far, in literature, about the composition of the intestinal microbiota in patients with epilepsy. To check for bacterial influence on epilepsy development and treatment, during my Ph.D. I’ve analyzed 3 datasets of epileptic patients: 1) a dataset of children admitted to the hospital after their first seizure and sampled 2 more times, after 4 and 12 months subjected to drug monotherapy; 2) a dataset of adults that are using daily for more than 2 years the most common drug available with no more seizures; 3) a dataset of girls affected by the Rett syndrome, a genetic disorder that comprises impairments in language and coordination and repetitive movements. Complications of Rett syndrome patients include seizures, and the patients take the same drugs as the epileptic dataset. This cohort has been added to the project as a negative control, as opposed to the healthy controls enrolled. This diverse dataset set the stage for speculations on the microbiota changes and development during epilepsy onset and constant drug therapy in terms of bacterial diversity, composition, and interaction. The bacterial DNA extracted from the fecal samples collected from the patients has been sequenced through next-generation sequencing techniques on a MiSeq Illumina platform, and the resulting reads have been analyzed through a bioinformatic pipeline using the QIIME software, R, and Matlab. VBiodiversity was observed to be decreased in pathological conditions and along with the drug assumption in all comparisons. Of particular interest is the peak of Akkermansiaceae in the children’s drug-naive gut microbiota. The children showed an interesting intra-individual signature and a diversity separation over pharmacological treatment time, possibly related to a reduced abundance of beneficial groups such as Faecalibacterium. From the co-abundance network analysis, the opposite trend of the group of genera related to Subdoligranulum and the ones related to Bacteroides have emerged. The adults in stable drug therapy showed some trends similar to the Rett patients, as if, despite the different clinical picture, the gut bacterial community was comparable and similar in terms of biodiversity. The feature selection analysis performed on the drug-naive and the drug-assuming patients confirmed the families Akkermansiaceae and Christensenellaceae to be relevant for the bacterial composition alteration during the pharmacological therapy. The separation of Gram-positive and Gram-negative bacteria showed an interesting change of the fractions with a trend in favor of the Gram-negative: the different protective layers could influence or be influenced by the drug assumption. The results shown in this thesis highlight the existence of differences in terms of general microbial diversity and taxonomy. Further studies carried out on a larger number of patients at different times and dise