Mesozooplankton is a key component of the ocean, regulating global processes such as the carbon pump, and ensuring energy transfer from lower to higher trophic levels. Yet, despite the importance of understanding mesozooplankton diversity, distribution and connectivity at global scale to predict the impact of climate change in marine ecosystems, there is still fragmented knowledge. To fill this gap, we applied DNA metabarcoding to mesozooplankton samples collected during the Malaspina-2010 circumnavigation expedition across temperate and tropical oceans from the surface to bathypelagic depths. By conducting a hidden diversity analysis, we highlight the still scarce knowledge on global mesozooplankton diversity and identify the Indian Ocean and the deep sea as the most understudied areas. By analysing mesozooplankton community spatial distribution, we confirm global biogeographical patterns across the temperate to tropical oceans both in the vertical and horizontal gradients. Additionally, we reveal a consistent increase in mesozooplankton beta-diversity with depth, indicating reduced connectivity at deeper layers, and identify a water mass type-mediated structuring of bathypelagic communities, instead of an oceanic basin-mediated as observed at upper layers. This suggests limited dispersal at deep ocean layers, most likely due to weaker currents and lower mixing of water mass types. Overall, our work supports the neutral theory of biodiversity and thus the importance of oceanic currents and barriers in dispersal in shaping global plankton communities, and provides key knowledge for predicting the impact of climate change in the deep-sea.