Engineering multi-agent systems (MAS) is known to be a complex task. One of the reasons lays in the complexity to combine multiple concerns that a MAS has to address, such as system functionality, coordination, robustness, etc. A well-recognized approach to manage system complexity is the use of self-adaptation (SA). Self-adaptation extends a system with support to monitor and adapt itself to realize a concern of interest (optimization, fault-tolerance, etc.). We present SA-MAS, an architectural approach that integrates MAS with SA. We present a reference model for SA-MAS and illustrate it with an excerpt from our research.