serious infections including tuberculosis, endocarditis, osteomyelitis, necrotizing pneumonia, and sepsis. [1] Treatment often requires long-term and intensive antibiotics administration; however, treatment failure and relapse are unfortunately common. [2] As we currently understand it, the major reasons for the failure of clinical therapy to eradicate intracellular bacteria include: i) poor cellular membrane penetration, suboptimal intracellular accumulation, and short retention of antibiotics; [3] ii) diminished antibacterial activity of antibiotics because of the harsh acidic and hydrolytic environment within phagolysosomes; [4] iii) intracellular bacteria being in a dormant state and tolerance of otherwise lethal concentration of antibiotics; [5] and iv) bacteria escape from phagolysosomes and hide in privileged intracellular compartments that evade the bactericidal actions of antibiotics. [6] At later timepoints, potentially after the cessation of therapy, the bacteria may then proliferate resulting in the apoptosis and autophagy of the cells. The evasive bacteria re-enter the circulation or re-infect local tissues. [7] As such, the infected cells have been likened to "Trojan horses" that protect bacteria with later dissemination of the infection into deeper tissues. [8] Drug delivery systems (DDSs) have shown increasing potential for the treatment of intracellular bacterial infection. [9] The Intracellular bacteria in latent or dormant states tolerate high-dose antibiotics. Fighting against these opportunistic bacteria has been a long-standing challenge. Herein, the design of a cascade-targeting drug delivery system (DDS) that can sequentially target macrophages and intracellular bacteria, exhibiting on-site drug delivery, is reported. The DDS is fabricated by encapsulating rifampicin (Rif ) into mannose-decorated poly(α-N-acryloylphenylalanine)-block-poly(β-N-acryloyl-d-aminoalanine) nanoparticles, denoted as Rif@FAM NPs. The mannose units on Rif@FAM NPs guide the initial macrophage-specific uptake and intracellular accumulation. After the uptake, the detachment of mannose in acidic phagolysosome via Schiff base cleavage exposes the d-aminoalanine moieties, which subsequently steer the NPs to escape from lysosomes and target intracellular bacteria through peptidoglycan-specific binding, as evidenced by the in situ/ex situ co-localization using confocal, flow cytometry, and transmission electron microscopy. Through the on-site Rif delivery, Rif@FAM NPs show superior in vitro and in vivo elimination efficiency than the control groups of free Rif or the DDSs lacking the macrophages-or bacteria-targeting moieties. Furthermore, Rif@FAM NPs remodel the innate immune response of the infected macrophages by upregulating M1/M2 polarization, resulting in a reinforced antibacterial capacity. Therefore, this biocompatible DDS enabling macrophages and bacteria targeting in a cascade manner provides a new outlook for the therapy of intracellular pathogen infection.