In-stream nitrate removal capacity may be used as a proxy for the ecosystem service of water quality regulation. It is well known that this natural function is driven by abiotic and biotic factors in running water environments. With regard to biotic drivers, most of the literature focuses on the microbial community influences, but there has been very little emphasis on the relationship with the benthic macroinvertebrate community. Since this community feeds on microbial assemblages (autotrophic and/or heterotrophic biofilms) that live on the streambed and in the hyporheic zone of the river, macroinvertebrates also have the potential to influence nitrate removal via its influences on microbiological processes. 2. The objective of this study was to examine the potential relationship between the macroinvertebrate communities and nitrate removal. A dataset of in-stream nitrate removal rates measured in nine-third-order streams was analysed. The simultaneous influences of abiotic (hydromorphological, physical and chemical characteristics) and biotic (biofilm and macroinvertebrate) drivers were examined and together explained 56% of the in-stream nitrate removal variance. An analysis of the independent contributions of each driver showed that abiotic drivers (e.g. ammonium, dissolved organic carbon, temperature and transient zone) contributed 40% of this nitrate removal variance, whereas the macroinvertebrate community contributed 39%. 3. The potential relationship between macroinvertebrates and nitrate removal was subsequently explored using trait-based approaches of the macroinvertebrate community. This method allows for the selection of trait modalities assuming a top-down control of microbial communities by macroinvertebrates, with in-stream abiotic conditions correlated with nitrate removal (assuming that environmental conditions affect macroinvertebrate community composition). 4. The main trait modalities positively correlated with nitrate removal were scraper (feeding habit), flagstones/boulders/cobbles/pebbles (substrate preference), crawler and interstitial (locomotion) and detritus (food). The main modalities negatively correlated with nitrate removal were silt and mud with microphytes (as substrate preference), and with fine sediment with microorganisms, and dead animals (as food sources). These results agreed with the hypothesis of top-down control and enhanced understanding of the influence of hydromorphological factors on nitrate removal. 5. This study highlights the involvement of the macroinvertebrate community in in-stream nitrate processing, and demonstrates the usefulness of applying a functional approach to explain relationships between biodiversity and ecosystem function.