Enteroids are an emerging research tool in the study of inflammatory bowel diseases such as necrotizing enterocolitis (NEC). They are traditionally grown in the basolateral-out (BO) conformation, where the apical surface of the epithelial cell faces the inner lumen. In this model, access to the luminal surface of enteroids for treatment and experimentation is challenging, which limits the ability to study host-pathogen interactions. To circumvent this, a neonatal apical-out (AO) model for necrotizing enterocolitis was created. Since intestinal epithelial cell permeability changes are pathognomonic for NEC, this protocol outlines using lucifer yellow (LY) as a marker of paracellular permeability. LY traverses the intestinal epithelial barrier via all three major paracellular pathways: pore, leak, and unrestricted. Using LY in an AO model allows for a broader study of permeability in NEC. Following IRB approval and parental consent, surgical samples of intestinal tissue were collected from human preterm neonates. Intestinal stem cells were harvested via crypt isolation and used to grow enteroids. Enteroids were grown to maturity and then transformed AO or left in BO conformation. These were either not treated (control) or were treated with lipopolysaccharide (LPS) and subjected to hypoxic conditions for the induction of in vitro NEC. LY was used to assess for permeability. Immunofluorescent staining of the apical protein zonula occludens-1 and basolateral protein β-catenin confirmed AO conformation. Both AO and BO enteroids treated with LPS and hypoxia demonstrated significantly increased paracellular permeability compared to controls. Both AO and BO enteroids showed increased uptake of LY into the lumen of the treated enteroids compared to controls. The utilization of LY in an AO enteroid model allows for the investigation of all three major pathways of paracellular permeability. It additionally allows for the investigation of host-pathogen interactions and how this may affect permeability compared to the BO enteroid model.