This paper considers the problem of designing urban mobility hubs by modeling them as service-integrated hub-line location problems. We integrate traditional network design decisions (where to place and how to connect the mobility hubs) with microscopic decisions regarding the services to offer at each located hub. The proposed model assumes that demand is elastic not only to the travel times, but also to the services provided at the entrance/egress points in the network, therefore providing more sensitive and realistic modeling capabilities. The proposed model aims to maximize the overall time savings of the demand captured by the service-enhanced hub-line. We propose a mixed-integer programming model for this problem, where demand elasticity is incorporated via gravity models. We conduct a case study using a real transit network, namely that of the metropolitan City of Montreal, Canada. We show that the proposed framework leads in certain cases to hub-lines whose topologies cannot be derived from classical hub-line network design models, showcasing the relevance of integrating the service location decisions and demand elasticity to the enhanced service levels in the planning of a hub-line.