Ductile fracture in steel is governed by the micromechanics of void nucleation, growth, and coalescence. The Gurson model (1977) effectively captures these mechanisms while maintaining computational efficiency. However, while the model performs well under high stress triaxiality, it is unable to predict damage correctly for low or negative triaxialities unless void nucleation under shear loading is explicitly considered [1]. In addition, the hot rolling process of structural steel profiles introduces microstructural anisotropy in the specimens due to crystallographic texture, segregation-induced banding, and the alignment of non-metallic inclusions. These elongated inclusions accelerate void coalescence in specimens tested perpendicular to the rolling direction [2]. To obtain a deeper understanding of the ductile fracture behaviour of hot rolled steel at low triaxialities and different loading angles, a combined approach of experimental testing and numerical modelling was used. Axisymmetric tensile tests were performed in seven orientations to characterise the anisotropic material properties, while single-bolt shear-out tests were performed to evaluate the lode dependence of the fracture strain. The experimental data were used to develop and validate a numerical model employing a shear-modified Gurson model extended with an orientation-dependent void coalescence. The results show that although the microstructural anisotropy has a negligible effect on the yield and tensile strength, it significantly influences the fracture strain. In addition, numerical simulations show that the use of a shear modified Gurson model enables accurate prediction of shear-out failure in a bolted joint connection.