Pre- and protostellar cores represent the earliest stage of the formation of a star. This phases are crucial for the future evolution of the star, as its final mass and the initial composition of the proto-planetary disk that may eventually form planet will be determined during these phases. In the past years, much progress has been done in our understanding of the chemical structure of these objects, thanks to the dramatic increase of the sensitivity of millimiter and sub-millimeter ground based telescopes. In fact, it is now possible to use chemistry as a tool to constrain both physical and chemical characteristics of these objects. However, most studies so far have used single dish observations with typical resolutions of a few tens of arc seconds, and the physical and chemical structures if cores on smaller scales remains poorly know. Here we propose to carry-out a study the physical and chemical properties of a large sample of Class 0 protostars, using both an observational and theorical/modelling approach. Our observations are part of an extensive survey of the line and continuum emission from young protostars with the Plateau de Bure interferometer at sub-arcsecond resolution. Follow-up observations with ALMA and NOEMA will be carried-out. We propose to develop a new chemo-dynamical model combining the results of state-of-the art MHD simulations of core collapse with a complete chemistry network and a radiative transfer model. The direct comparison between sub-arcsecond resolution observations and the predictions our chemo-dynamical model is expected to bring important constraints on the formation and evolution of pre- and protostellar cores, and in turn on star formation theories.