Discovering spatial co-location patterns with both high prevalence and utility is essential for utility-aware spatial analysis. However, existing High Utility Co-location Pattern Mining (HUCPM) methods assume all utilities are positive, limiting their applicability. To overcome this, we propose PN-HUCP (Positive-Negative Utility High Utility Co-location Pattern miner), a novel framework that supports both positive and negative utilities. PN-HUCP introduces a pattern utility index and four effective pruning strategies to enhance mining efficiency. Experiments on synthetic and real-world datasets demonstrate that PN-HUCP accurately identifies all qualified patterns, reduces candidates by 97.38% on average, and achieves superior performance compared to state-of-the-art methods.