Dara Azuma ^1* Yvette Penner ² Tomoko Kaneko-Tarui ³ Taysir Mahmoud ³ Janis L Breeze ⁴ Angie Rodday ⁵ Perrie O'tierney-Ginn ³ Jill L Maron ⁶

• ¹ Virginia Commonwealth University, Richmond, Virginia, United States
• ² Boston Children's Hospital, Boston, United States
• ³ Tufts Medical Center- Mother Infant Research Institute, Boston, United States
• ⁴ Dalhousie University, Halifax, Nova Scotia, Canada
• ⁵ Tufts University, Medford, Massachusetts, United States
• ⁶ Department of Pediatrics, Warren Alpert Medical School, Brown University, Providence, Rhode Island, United States

Introduction: Infants of diabetic mothers (IDMs) may exhibit decreased oral intake, requiring nasogastric feedings and prolonged hospitalization. The objective of this study was to explore whether saliva serves as an informative biofluid for detecting expression of hunger signaling and energy homeostasis modulator genes and to perform exploratory analyses examining expression profiles, body composition, and feeding outcomes in late preterm and term IDMs and infants born to mothers with normoglycemia during pregnancy. Methods: In this prospective cohort pilot study, infants born at ≥ 35 weeks' gestation to mothers with gestational or type II diabetes (IDM cohort) and normoglycemic mothers (control cohort) were recruited. The presence of known hunger signaling genes: 5'AMP-activated protein kinase (PRKAA2) and neuropeptide Y2 receptor (NPY2R); adipokines: leptin (LEP) and adiponectin (ADIPOQ); and energy homeostasis regulators: ghrelin (GHRL) and proopiomelanocortin (POMC) in neonatal saliva was determined with RT-qPCR and compared between cohorts. Body composition was assessed via skinfold measurements and compared between cohorts. Feeding outcomes were recorded. Exploratory analyses were performed examining associations between infant body composition, energy homeostasis and hunger signaling gene expression.Results: Twenty-three infants in the IDM cohort and 22 infants in the control cohort were recruited. LEP and ADIPOQ were not reliably detected in neonatal saliva in either cohort. PRKAA2, GHRL and NPY2R were less likely to be detected in the IDM cohort, whereas POMC was more likely to be detected in the IDM cohort. Infants in the IDM cohort had greater adiposity compared to infants in the normoglycemia cohort. Only 3 IDMs had documented poor feeding; no infant in the control group struggled to feed. In exploring associations between hunger signaling gene expression with energy homeostasis gene expression and body composition, the odds of detecting salivary NPY2R expression decreased as fat mass increased, and the odds of detecting PRKAA2 expression increased in the presence of GHRL expression.Discussion: Non-invasive assessment of hunger signaling and energy homeostasis gene expression is possible through neonatal salivary analysis. This pilot study lays the foundation for a larger scale study to further investigate the link between in utero exposure to diabetes with body composition and regulation of appetite.