Antimicrobial resistance (AMR) is a global healthcare emergency that is predicted to kill approximately 10 million people annually by 2050 and cost the global economy ~$100 trillion. The World Health Organisation therefore emphasises the need of strengthening AMR knowledge and improving AMR surveillance to reduce the burden of AMR. However, AMR surveillance is complicated as existing AMR genes are not attributable to all phenotypic antibiotic resistance. Using a combination of forced evolution and metabolic rescue experiments we have identified a novel mechanism of resistance to sulfamethoxazole (an inhibitor of bacterial folate synthesis) in Group A Streptococcus, a pathogen that causes Strep throat, skin sores and potentially autoimmune diseases (e.g., rheumatic heart disease)1. This novel sulfamethoxazole resistance is mediated by a substrate-binding protein (ThfT) of an ECF transporter that expands the substrate profile of an existing transporter system to include tetrahydrofolate and other related folate end products. In the context of an infection, ThfT acquires folate metabolites directly from the human host to bypass the inhibitory action of sulfamethoxazole and this resistance remains undetectable by routine antibiotic susceptibility testing methods in pathology laboratories as in vitro culture media does not contain adequate folate metabolites. thfT is likely acquired by horizontal gene transfer from related Streptococcus species (Streptococcus dysgalactiae subsp. equisimilis) and GAS strains with thfT were able to grow in the presence >5000 µg/ml of sulfamethoxazole. We propose that ThfT-mediated resistance constitutes a new paradigm of antibiotic resistance and could be used as a model to investigate host metabolite dependent antibiotic resistance in other medically important bacterial pathogens. Additionally, our study also highlights the need to understand the activity of the antibiotics in the context of the infections they are designed to treat.