The importance of conceptual modeling in software engineering is constantly growing along with the complexity of business domains. In order to handle this increasing complexity, it is essential to use modeling notations that provide a high level of abstraction and expressiveness. For this reason, we specifically focus on ontology-driven conceptual modeling and OntoUML notation in our research. In this paper, we present the main ideas behind our novel approach to OntoUML model instantiation using multi-labeled property graph databases, such as Neo4j. Our approach should not only serve to ensure data integrity in graph databases but also provide domain experts with a convenient way to verify the correctness and completeness of their conceptual models. Although there already exist approaches to instantiate OntoUML models using relational databases, our research shows that multi-labeled property graph databases are better suited for this purpose as their versatile data model better correlates to the essential principles of OntoUML.