The development of HIV-1 integrase (INT) inhibitors has been hampered by incomplete structural and mechanistic information. Despite the efforts made to overcome these limitations, only one compound has been approved for clinical use so far. In this work, we have used all experimental information available for INT and similar enzymes, to build a model of the holo-integrase:DNA complex that includes an entire central core domain, a ssDNA GCAGT substrate, and two magnesium ions. Subsequently, we used a large array of computational techniques, which included molecular dynamics, thermodynamic integration, and high-level quantum mechanics/molecular mechanics (QM/MM) calculations to study the possible pathways for the mechanism of 3' end processing catalyzed by INT. We found that the only viable mechanism to hydrolyze the DNA substrate is a nucleophilic attack of an active site water molecule to the phosphorus atom of the scissile phosphoester bond, with the attacking water being simultaneously deprotonated by an Mg2+-bound hydroxide ion. The unstable leaving oxoanion is protonated by an Mg2+-bound water molecule within the same elementary reaction step. This reaction has an activation free energy of 15.4 kcal/mol, well within the limits imposed by the experimental turnover. This work significantly improves the fundamental knowledge on the integrase chemistry. It can also contribute to the discovery of leads against HIV-1 infection as it provides, for the first time, accurate transition states structures that can be successfully used as templates for high-throughput screening of new INT inhibitors.