Heavy metals are some of the most persistent and potent anthropogenic environmental contaminants. Although heavy metals may compromise microbial communities and soil fertility, it is challenging to causally link microbial responses to heavy metals due to various confounding factors, including correlated soil physicochemistry or nutrient availability. A solution is to investigate whether tolerance to the pollutant has been induced, called Pollution Induced Community Tolerance (PICT). In this study, we investigated soil microbial responses to a century-old gradient of metal ore pollution in an otherwise pristine subarctic birch forest generated by a railway source of iron ore transportation. To do this, we determined microbial biomass, growth, and respiration rates, and bacterial tolerance to Zn and Cu in replicated distance transects (1 m–4 km) perpendicular to the railway. Microbial biomass, growth and respiration rates were stable across the pollution gradient. The microbial community structure could be distinguished between sampled distances, but most of the variation was explained by soil pH differences, and it did not align with distance from the railroad pollution source. Bacterial tolerance to Zn and Cu started from background levels at 4 km distance from the pollution source, and remained at background levels for Cu throughout the gradient. Yet, bacterial tolerance to Zn increased 10-fold 100 m from the railway source. Our results show that the microbial community structure, size and performance remained unaffected by the metal ore exposure, suggesting no impact on ecosystem functioning. An induced bacterial Zn-tolerance demonstrated that pristine soil microbial communities had been contaminated by metal pollution derived from iron ore transport.