Modern distributed systems comprise of a number of loosely-coupled services, which collaborate, interact and share data to accomplish a task. This paradigm is currently used by many systems such as Service Oriented Architecture, Cloud Computing, Product Lifecycle Management, Pervasive Healthcare, Digital Supply Chains, Digital Rights Management, etc. An interaction can involve multiple parties, where each party generates, shares, uses and interacts with the data. Existing technologies for secure data sharing in point-to-point or client-server paradigms don't work in the distributed environment because of the involvement of multiple parties in an interaction. The main security challenge is that the data owner has no control and visibility on interactions beyond its trust domain, so it is difficult to understand and track data dissemination to unknown or untrusted parties. The interactions beyond the trust domain of data owner may share data to unauthorized parties and violate owner's policies and the owner has no way of knowing if a violation occurred. Such interactions introduce new security challenges not present in the traditional systems, where the focus is to ensure security within a single domain. Thus it is imperative to have a new solution that supports secure data dissemination in distributed interactions and protects data according to its owner's policies. In this poster, we describe a data-centric approach for end-to-end secure data dissemination. This approach transforms passive data into an active entity that is able to protect itself. It enables dynamic data dissemination decisions and protects data throughout its life cycle. The granularity of the data being shared with a party is determined by the respective dissemination policy of the data owner. This enables sustainability of data confidentiality outside the owner's trust domain.