Patofisiologia do ZIKA vírus em glândulas salivares e detecção biofotônica do ZIKA vírus na saliva associado com algoritmo de inteligência artificial

Abstract

The Zika virus (ZIKV) belongs to the Flaviviridae family, therefore, it is related from the evolutionary point of view to other arboviruses transmitted by mosquitoes. The natural cycle of ZIKV transmission in Africa mainly involves wild vectors of the genus Aedes and monkeys. In late 2013, an epidemic outside Africa was reported in French Polynesia with more than 35,000 cases. This event alerted health officials to the potential for a pandemic spread of this virus which has also been identified in New Caledonia, the Cook Islands, and Easter Island. In 2015, the circulation of the ZIKV in Northeast Brazil was confirmed. Recently, the Ministry of Health of Brazil published a letter to confirm the presence of ZIKV in eight states, in the Northeast and Southeast regions of Brazil. Recent findings show the presence of ZIKV in blood, semen, urine, and saliva, suggesting that the transmission could also be through these additional fluids. Saliva is a non-invasive, painless biofluid, with a reduced collection cost compared to blood and can be highly accurate using analytical platforms appropriate for each disease. However, evidence of ZIKV infection in salivary glands and its potential effect on salivary diagnosis is still incipient literature. In this context, the objective of the thesis was divided into 3 chapters: 1) present a physiological mechanism of ZIKV entry into salivary gland cells and its potential impact on salivary diagnosis; 2) evaluate the molecular composition of the submandibular glands in an animal model of ZIKV; and 3) assess the discriminatory discrimination ability of ZKV in different saliva products using a sustainable, reagent-free, rapid, and non-invasive biophonic platform. In the first chapter, it was revisited the expression of mRNA and proteins in salivary glands related to the molecular mechanism of ZIKV infection in salivary glands. Among them are type C lectin receptors and phosphatidylserine receptors (PS type TIM1, TIM3, TIM4, TYRO3, and AXL). From a pathophysiological perspective related to diagnosis, the salivary glands can allow the entry of ZIKV through the aforementioned transporters, allow the replication of ZIKV and its secretion to the acinar and duct lumen until it reaches the oral cavity with saliva, which is essential to develop appropriate strategies to perform diagnostic platforms for early detection of ZIKV in saliva. In the second chapter, it was developed an animal model of ZIKV infection using two-month-old male C57/BL6 knockout mice for the interferon-gamma gene. ZIKV infection was performed by intradermal administration with ZIKV (50 µL, 1 x 105 PFU) and the control received vehicle (50 µL). To confirm ZIKV infection in this animal model, ZIKV RNA replication was evaluated in the spleen of mice. It was observed that the presence of collagen reduced (p < 0.05) and nucleic acids increased (p < 0.05) in the submandibular glands of ZIKV-infected mice. In the third chapter, it was identified that a machine learning algorithm based on LDA discriminated infected saliva with 10⁴ PFU/mL, which is similar to that found clinically in ZIKV infection, with 80.5% accuracy (sensitivity: 77.7% and specificity: 83.3%) from non-infected saliva. In this way, the results demonstrate a potential application of this sustainable biophotonic platform without the use of reagents coupled with machine learning algorithms to detect ZIKV in saliva. Together, the review on the expression of transporters for ZIKV entry into salivary gland cells allowed an understanding of the pathophysiology of ZIKV in the major and minor salivary glands and its effects on oral health. The detection of alterations in the expression of collagen and nucleic acids in an animal model of ZIKV reinforces the importance of the pathophysiology of ZIKV in tissues of the oral cavity, and also that further studies should be directed to the effects of ZIKV in salivary glands. Furthermore, we demonstrate that a sustainable biophotonic platform using infrared spectra of saliva coupled with an artificial intelligence algorithm can detect ZIKV at concentrations similar to those found clinically.