Well-Characterised Time-Gated Detector Photon Flux Resolves the Ultrastructure of DNA-Damage Nuclear Bodies with G-STED Nanoscopy

Widely spread commercial STED nanoscopes, and low-cost custom-built systems are generally equipped with continuous-wave STED lasers (cwSTED) combined with pulsed excitation lasers (peLaser), and time-gated detection. This modality, termed gated-STED (gSTED), permits using lower cwSTED powers reducing background noise, at the expense of low intensity images. gSTED critically depends on the photon-counting detectors sensitivity, and photon-flux optimization to the temporal gate. The commercial system employed in this work utilise an electronic trigger-box to synchronise the peLaser with time-gate hybrid detectors (gHyD). Unfortunately, commercial gSTED set-ups do not provide an accurate detection temporal gate characterization. Hence, users blindly select the time-gated window with none a priori knowledge of neither the shape, nor the delay relative to the peWLL. As a result, often the acquisition of sharp, highly-resolved images turns into seemingly random events. Here, we report how to measure the gate shape in our gSTED microscope. Our gate starts at 2.5ns and last 4.5ns when exciting at 488nm, depleting at 592nm, and with both notch filters. The effective gate shape is highly affected by the fluorescence emission and detection, as well as the optical elements in the pathway. For instance, neither a detection window from 500nm and above 560nm, nor short lifetime dyes yield sharp, highly-resolved images. Furthermore, a thorough study on how different peLaser and cwSTED power ratios, and different time-gates impact the lateral resolution and image signal-to-noise allowed generating resolution maps showing a priori knowledge to best gSTED performance. Our method allowed resolving the nuclear bodies (NB) ultrastructure, a previously unresolved sub-nuclear DNA assembly in human cells. They spontaneously appear in ∼10% of G1/early S phase cells, and it is suggested to be DNA damage sheltering centre for transgenerational DNA lesions originated in the previous S phase. Our method revealed the NB organisation in high detail, showing that NB contains multiple chromatin looped structures in the periphery. To gain further details of the NB inner core, we extended the chromatin by pre-extracting cells prior to staining. Compared to the previous condition, we observe chromatin fibre bundles of approximate 15-40nm diameter, and can now resolve a higher-order chromatin organisation reminiscent of a molecular plait-like sub-structure. In conclusion, a well-characterised time-gate, and cwSTED/peWLL power ratio in commercial gSTED nanoscopes allows resolving the NB ultra-structure with a consistent resolution (within ∼15nm) and possibly other biologically-relevant structures.