With the rise of the city-regions development strategy, the impact of the spatial structure in city-regions on environmental pollution becomes an increasingly important issue related to sustainable regional development. Although previous studies have paid attention to spatial structure affecting environmental quality, there is a lack of analyzing the reduction effect of industrial pollution emission from the perspective of polycentricity and monocentricity. In order to provide the evidence from industrial production activities to test the relationship between spatial structure and industrial pollution, this study extends the production density model by treating industrial pollution emissions as non-expected output and introducing the spatial structure factor into the model, thus revealing the non-linear interactive effects of spatial structure and agglomeration on industrial pollution emissions, and exploring this emission reduction effect and its mechanisms using threshold regression model and mediation model. The empirical analysis shows that spatial structure matters and the results are as follows: Firstly, polycentricity is associated with lower industrial pollution emission intensity, but the emission reduction effect of polycentricity depends on regional agglomeration. Secondly, the higher the regional agglomeration, the greater the emission reduction effect. Thirdly, the heterogeneity test also indicates that the geographical compactness and unified administration of city-regions strengthen this emission reduction effect. Fourthly, the mediating effect test verifies the three mechanisms proposed in this study, among which energy utilization efficiency effect plays a major role. The governments are expected to promote the transformation of the spatial structure from a monocentricity to polycentricity based on the level of regional agglomeration to ensure that the expected emission reduction effect can be achieved.