A frontier challenge in bioscience is to obtain deep mechanistic understanding from atomic resolution, time-resolved structures of macromolecules engaged in function. To this end, we have developed sample-efficient delivery and reaction initiation strategies that use room temperature microcrystal slurries and serial crystallography methods for time-resolved studies. Our overriding hypothesis is that microcrystal enzyme-substrate complexes will form many times faster than the average enzyme catalytic rate, which is supported by increasing numbers of observations. However, interpreting electron density maps from reaction cycle intermediates can be challenging, especially when mixtures of species are often present in the data. Therefore, to help reduce ambiguity we have also pioneered strategies to simultaneously collect time-resolved serial crystallography (tr-SMX/SFX) diffraction data in the forward direction, and X-ray emission spectroscopy (tr-XES) data at ~ 90º, using either X-ray free electron laser (XFEL) sources, or synchrotron sources such as Diamond Light Source in Oxfordshire UK. The resulting atomic and electronic structures are fully correlated and have been applied to a range of metalloenzymes. For instance, isopenicillin N synthase (IPNS) uses nonheme iron to catalyse the O2-dependent conversion of its tripeptide substrate delta-(L-alpha-aminoadipoyl)-L-cysteinyl-D-valine (ACV) into isopenicillin N (IPN, the precursor of all penicillin/cephalosporin beta-lactam antibiotics). The unique four electron oxidation reaction leading to the beta-lactam bicyclic ring proceeds via two high-valent iron species, an Fe(III)-superoxo and a high-spin Fe(IV)=O oxyferryl species. These enable two sequential C-H bond cleavage steps that each exhibit large kinetic isotope effects (KIE). Our recent tr-SFX and tr-XES studies have characterised the Fe(III)-superoxo species and revealed unexpected, correlated motions throughout the whole protein caused by O2 binding.