In this work, we propose MobiTrust, a game theoretic pricing scheme, to address the problem of ensuring trust in the edge-enabled mobile sensor-cloud (MSC) architecture for vehicular Internet-of-Things (IoT). In the edge-enabled MSC, a sensor-cloud service provider (SCSP) utilises multiple sensor-equipped vehicular IoT-based devices belonging to different owners and multiple edge devices deployed by different edge infrastructure providers (IPs) to provision Sensors-as-a-Service (Se-aaS). Since these entities are revenue-driven and they depend on each other for successful service provisioning, there is a need to establish trust among them to ensure the sustainability of MSC. Existing works on sensor-cloud address the issue of ensuring trust among selfish device-owners. However, these works do not consider the presence of IPs and assume that the IoT devices are stationary in nature, which renders them unsuitable for MSC. Therefore, in this work, we aim to address this drawback of the existing literature by proposing MobiTrust, which is comprised of two components. First, we propose a mobility-aware trust model designed based on the Transferable Belief Model, which utilizes different types of evidence collected by the devices to estimate their trust values. Second, we propose a trust-based pricing and resource allocation scheme based on the Stackelberg game that aims to maximise the profits and satisfaction of the various stakeholders. We analyze the proposed scheme theoretically and examine the existence of equilibrium in the proposed game. Results of extensive simulations show that MobiTrust increases profits for the stakeholders and QoS by 9% and 11%, respectively, compared to benchmarks.