Security in Systems-of-Systems remains a major challenge due to their distributed, heterogeneous, and autonomous nature. Trusted Execution Environments have emerged as a promising hardware-based approach to address confidentiality, integrity, and trust in such complex ecosystems. However, the application of TEE within SoS lacks a systematic overview. This study conducts a Systematic Mapping Study to comprehensively assess the current state of TEE adoption in SoS, identifying primary challenges, outcomes, and evaluation approaches. A reproducible protocol was applied across four scientific databases, yielding 301 studies, of which 39 were selected for detailed analysis. Findings reveal that research predominantly addresses domains such as Smart Grids, Industrial IoT, and Smart Cities, focusing on requirements such as confidentiality, attestation, privacy, and isolation. Key challenges include performance overhead, hardware-level vulnerabilities, and integration complexity. Despite these obstacles, TEE technologies demonstrate transformative potential by enabling new architectural patterns for secure computation, privacy-preserving collaboration, and trustworthy system integration. The study also highlights underexplored application areas and proposes a research agenda that emphasizes lightweight TEEs, interoperability, and robust evaluation methodologies. Finally, this SMS identifies research gaps and proposes future directions to expand TEE adoption and improve scalability, interoperability, and empirical validation in SoS environments.