The mammalian Golgi consists of at least three biochemically distinct cisternae, cis-, medial- and trans (reviewed in Szul and Sztul, 2011; Day et al, 2013). The structure and function of the Golgi are tightly interconnected, such that proteins that are required for protein transport through the Golgi are often also required for the organization of the Golgi stacks, and vice versa (reviewed in Liu and Storrie, 2012; Liu and Storrie, 2015; Chia and Gleeson, 2014; Munro, 2011). Newly synthesized proteins from the ER and ERGIC are received at the cis face of the Golgi and flow through to the trans-Golgi before being released to the trans-Golgi network (TGN) for further secretion to the endolysosomal system, plasma membrane or extracellular region. Retrograde flow from the trans- to cis-cisternae moves endocytosed cargo from the extracellular region, the plasma membrane and the endolysosomal system back towards the ER. Intra-Golgi retrograde traffic also returns resident Golgi proteins to their appropriate cisternae, in this way facilitating cisternal remodeling or maturation (reviewed in Boncompain and Perez, 2013; Day et al, 2013). Intra-Golgi traffic in both directions is mediated by COPI carriers, with specificity of transport being determined at least in part by the complement of SNAREs, RABs and tethering proteins involved (reviewed in Szul and Sztul, 2011; Spang 2013; Willet et al, 2013; Chia and Gleeson, 2014).