Among infectious diseases, tuberculosis is the leading cause of death worldwide, and represents a serious threat, especially in developing countries. The protective effects of Bacillus Calmette-Guerin (BCG), the current vaccine against tuberculosis, have been related not only to specific induction of T-cell immunity, but also with the long-term epigenetic and metabolic reprogramming of the cells from the innate immune system through a process termed trained immunity. Here we show that MTBVAC, a live attenuated strain of Mycobacterium tuberculosis, safe and immunogenic against tuberculosis antigens in adults and newborns, is also able to generate trained immunity through the induction of glycolysis and glutaminolysis and the accumulation of histone methylation marks at the promoters of proinflammatory genes, facilitating an enhanced response after secondary challenge with non-related bacterial stimuli. Importantly, these findings in human primary myeloid cells are complemented by a strong MTBVAC-induced heterologous protection against a lethal challenge with Streptococcus pneumoniae in an experimental murine model of pneumonia. Author summary Mycobacterium tuberculosis has been causing infections in our species and our ancestors for at least thousands of years. Still today, the numbers of people affected by tuberculosis are alarming with more than 1,4 million deaths per year, representing the first cause of death by infectious disease worldwide. Despite immense research efforts, Bacille Calmette-Guerin (BCG), a vaccine based on a live attenuated form of Mycobacterium bovis that was developed one century ago, remains the only licensed vaccine against tuberculosis. Several independent works have shown that BCG induces protection not only against tuberculosis, but also against other infections through the induction of epigenetic and metabolic effects in the cells of the innate immune system, what has been termed trained immunity. Here we describe how MTBVAC, a new vaccine against tuberculosis based on a genetically modified form of the human pathogen Mycobacterium tuberculosis, that has shown similar safety profiles and improved protection than BCG in preclinical studies, is able to induce trained immunity in human cells and grant protection against a heterologous model of pneumococcal infection in mice. These results underline the potential of MTBVAC as a candidate for universal vaccination against tuberculosis or treatment against bladder cancer, representing a potential alternative to the current BCG vaccines.