BACKGROUND: Early detection of cancer biomarkers such as microRNA-21 (miR-21), a small RNA molecule, can facilitate earlier diagnosis and potentially lead to earlier treatment. The detection of trace miRNA in complicated cellular environments necessitates the construction of self-sustainable DNA circuitry boasting high signal gain and robust anti-interference capabilities. However, current self-sustainable DNA circuits suffer from complex designs and severe signal leakage. Therefore, it is essential to develop a highly efficient and reliable strategy to improve sensing performance and accurately detect trace biomolecules in intricate biological matrices. RESULTS: We engineered a general and high-performance self-sustainable DNA amplification (SDA) circuit for reliable bioimaging inside cells. The autocatalytic SDA system was composed of the catalytic DNA assembly (CDA) and the rolling circle amplification (RCA) module. Upon input of the initiator, it stimulated the self-driven cross-invasion of the CDA and RCA amplicon, facilitating the successive replication of initiator sequences and resulting in synergistically accelerated and exponential signal amplification, as systematically investigated by various experimental studies. Due to its highly efficient amplification capability and universal applicability, the self-driven SDA system enabled reliable determination of miR-21 in buffer and serum and achieved a low detection limit of 8.9 pM. As a powerful imaging strategy, the SDA circuit realized accurate miR-21 imaging within cells, highlighting its potential for clinical diagnosis. SIGNIFICANCE: Reciprocal reinforcement of the CDA and RCA amplifiers accelerates the entire reaction process, facilitating the generation of an exponentially amplified FRET signal for reliable detection of analytes. The proposed SDA strategy achieves the one-step determination of DNA or miRNA with simple design, high signal gain, low signal leakage, and single-base specificity, and furthermore enables reliable miRNA localization inside cells, highlighting its potential to monitor significant biomolecules and substantially expanding the toolkits for clinical diagnosis.