This chapter delves into the integration of Graph Neural Networks (GNNs) within Neural-Symbolic Computing (NSC), highlighting the symbiosis between machine learning and symbolic reasoning. We explore various GNN models, including Logic Tensor Networks, Pointer Networks, Graph Convolutional Networks, and Graph Attention Networks, emphasizing their role in enhancing neural-symbolic integration. The chapter illuminates the potential of GNNs in processing and interpreting graph-structured data, which is crucial for applications requiring complex relational understanding. We also discuss possible future directions for NSC, like unexplored use cases, EM-based end-to-end methods, and combining data from different sources. This shows that GNNs are becoming more useful and widespread in AI. The insights presented in this chapter underscore the transformative impact of GNNs in NSC, positioning them as pivotal tools in both theoretical and practical AI advancements.