MINSTED fluorescence localization and nanoscopy

Articles https://doi.org/10.1038/s41566-021-00774-2 MINSTED fluorescence localization and nanoscopy Michael Weber1,4, Marcel Leutenegger1,4, Stefan Stoldt1,2, Stefan Jakobs Alexey N. Butkevich 3 and Stefan W. Hell 1,3 ✉ , Tiberiu S. Mihaila 1,2 , 1 We introduce MINSTED, a fluorophore localization and super-resolution microscopy concept based on stimulated emission depletion (STED) that provides spatial precision and resolution down to the molecular scale. In MINSTED, the intensity minimum of the STED doughnut, and hence the point of minimal STED, serves as a movable reference coordinate for fluorophore localization. As the STED rate, the background and the required number of fluorescence detections are low compared with most other STED microscopy and localization methods, MINSTED entails substantially less fluorophore bleaching. In our implementation, 200–1,000 detections per fluorophore provide a localization precision of 1–3 nm in standard deviation, which in conjunction with independent single fluorophore switching translates to a ~100-fold improvement in far-field microscopy resolution over the diffraction limit. The performance of MINSTED nanoscopy is demonstrated by imaging the distribution of Mic60 proteins in the mitochondrial inner membrane of human cells. T o resolve fluorophores that are far closer than the diffraction limit, all lens-based fluorescence nanoscopy methods have to make adjacent fluorophores discernible during registration and identify their coordinates with high precision. The elegance of STED microscopy1,2 derives from the fact that both tasks are performed in one go by the doughnut-shaped STED beam. By confining the fluorescence ability to a sub-diffraction-sized region around its central minimum, the STED doughnut beam both singles out the fluorophores that happen to be located in this region and establishes their position. The fluorescence ability and therefore the region defined by the STED doughnut are well described by the effective point-spread-function (E-PSF) of the STED microscope3p , affiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Gaussian ffi of full-width-half-maximum (FWHM) d  λ=ð2NA 1 þ I=Is Þ. Here λ, NA, I and Is denote the wavelengthIof the STED beam, the numerical aperture of the lens, the focal peak intensity at the doughnut crest and the intensity that reduces the fluorescence ability by half, respectively. Thus, scanning the sample with co-aligned (typically sub-nanosecond pulsed) excitation and STED beams separates fluorophores that are further apart than d and also locates them with the standard deviation σE ≈ 0.42d. Interestingly, if d becomes as small as the fluorophore itself (1–2 nm), which is theoretically possible for I > 104Is, all fluorophores will be prevented from fluorescing except the one that happens to be located right at the central doughnut minimum. At this conceptual limit without background, detecting just a single photon per fluorophore renders a perfect image, because a single detection within a given time span verifies the presence of a fluorophore at a coordinate perfectly defined by the doughnut. No other super-resolution fluorescence concept can make emitted photons as informative as STED microscopy and its close derivatives4. Unfortunately, separating and locating the emitters in one go comes at a cost. Since fluorescence blocking by STED typically entails intensities of Is ≈ 1–10 MW cm−2, discerning fluorophores closer than d = 20 nm requires I > 100Is ≈ 0.1–1 GW cm−2. Apart from the fact that applying such intensities to excited fluorophores promotes bleaching, doughnut minima are rarely <0.01I in practice3 due to residual alignment errors and aberrations. For I > 100Is, this means that the intensity at the minimum exceeds Is, which also degrades the fluorescence probability at the targeted coordinate and thus the fluorophore separation at distances well below 20 nm. Here we introduce MINSTED nanoscopy, a STED-based super-resolution fluorescence microscopy method that can provide molecule-size (1–3 nm) spatial resolution. This breakthrough has become possible by not requiring the STED doughnut to separate fluorophores (at small distances); its role is rather to establish the fluorophore’s position. Although we give up some of the elegance of the original STED concept, we obtain a fluorescence microscopy method whose resolution can be tuned from the diffraction limit down to the size of the fluorophores themselves. Compared with most other advanced STED and super-resolution methods4, MINSTED nanoscopy and the pertinent MINSTED localization entail less bleaching and reach the molecular scale with much fewer detected photons than achieved by popular camera-based techniques. Results MINSTED principle. To separate fluorophores at nanometre distances, MINSTED nanoscopy employs fluorophores that are transferred from an inactive (off) to an active (on) state and back. In the active state the fluorophore can be optically excited and de-excited by stimulated emission as in the concept called protected STED5. However, in MINSTED nanoscopy only one fluorophore within a diffraction-limited region is switched on at any given time, meaning that its coordinate is initially unknown across diffraction length scales6,7. The subsequent localization with the STED beam is greatly facilitated by the fact that the central minimum of the doughnut defines a coordinate to which the unknown coordinate of the fluorophore can be related. In the subsequent text, we refer to the position of the doughnut minimum as the ‘doughnut position’. Since it can be steered with beam deflectors at sub-nanometre precision, the doughnut position can be used for finding the position of the fluorophore in a sample: the closer it is to the fluorophore, the lower is the STED probability and the more probable is fluorescent emission. Evidently, the doughnut position entailing minimal STED must be identical with the fluorophore coordinate, hence the name MINSTED. Department of NanoBiophotonics, Max Planck Institute for Biophysical Chemistry, Göttingen, Germany. 2Clinic of Neurology, University Medical Center Göttingen, Göttingen, Germany. 3Department of Optical Nanoscopy, Max Planck Institute for Medical Research, Heidelberg, Germany. 4These authors contributed equally: Michael Weber, Marcel Leutenegger. ✉e-mail: 1 Nature Photonics | VOL 15 | May 2021 | 361–366 | www.nature.com/naturephotonics 361 Articles In contrast to the related concept called MINFLUX8, searching for the doughnut position with minimal STED is tantamount to searching for the position where the fluorescence is maximal. Yet this does not imply maximizing emission per se. First, the absolute emission rate is freely adjustable via the excitation beam power. Second, and more importantly, placing the doughnut minimum on top of the fluorophore to maximize the emission is neither required nor desired. Since the E-PSF is a Gaussian function, moving the E-PSF maximum in close proximity to the fluo (...truncated)