METHOD AND DEVICE FOR LIGHT MICROSCOPIC MULTISCALE RECORDING OF BIOLOGICAL SAMPLES

Description

TECHNICAL FIELD OF THE INVENTION

[0001] The invention relates to methods of light-optical microscopy, in particular to methods of super-resolved, non-diffraction-limited optical fluorescence microscopy on biological samples. It also relates to combination methods in which non-diffraction-limited optical fluorescence microscopy methods form a component. Furthermore, it also relates to method combinations with method components that are light-optical but not fluorescence microscopy, as well as to those with method components that are not light-optical. The invention further relates to corresponding devices. STATE OF THE ART

[0002] German patent DE 10 2018 126 232 B3 discloses a scanning microscope comprising multiple light sources. The light sources can be narrowband lasers or white light lasers. The light sources are coupled individually or jointly into a common illumination beam path using a dichroic mirror. The coupling occurs in a section through which detection light radiates in the opposite direction. The set of different light sources can be intended for different purposes; for example, the individual light sources in the set can emit excitation light for confocal microscopy, for MINFLUX nanoscopy, or for STED microscopy of various dyes. A light source can also serve as a light source for fluorescence-preventing light, particularly STED light.In this case, the microscope must also have a beam-shaping device that adjusts the wavefront of the illuminating light so that a zero point of intensity of the fluorescence-preventing light, surrounded by intensity maxima, forms at the focal point of the objective. A corresponding beam-shaping device must also be present in MINFLUX nanoscopy. The patent is directed at a changing device by means of which a dichroic mirror assigned to a light source is exchanged for a plane-parallel plate, which imposes the same parallel offset on the beam path as the dichroic mirror, when the light source in question is not required for illumination. This ensures, on the one hand, that light emitted from the sample at the wavelength of the unused light source can also be detected, and, on the other hand, that the alignment of the beam paths to each other is maintained.The scanning microscope also has an objective lens through which the illumination light is focused onto a sample, and a scanner that shifts the light intensity distribution of the illumination light relative to the sample and descans the detection light. Furthermore, it has an adjustable wavelength-selective bandpass filter in the detection beam path, which consists of an adjustable short-pass filter and an adjustable long-pass filter. The light passing through the filter is fed to a detector through a detection light output, while the reflected light is fed to another output. The reflected light can then be fed to another adjustable wavelength-selective bandpass filter, meaning the filters can be cascaded. Using the filter, the spectrum of the detection light can be analyzed by continuously tuning the filter.

[0003] German patent application DE 10 2010 015 915 A1 discloses a method for determining excitation spectra and emission spectra of fluorescent dyes in a sample. This method is carried out using a microscope. The microscope is equipped with a plurality of detectors, which alternately detect light of different wavelength bands. All wavelength bands can be spectrally shifted, but even after the shift, the detectors continue to detect light of different wavelength bands. The microscope also features a laser whose emission wavelength can be spectrally shifted.To determine the excitation spectrum, the sample is first excited at a first wavelength, and the fluorescence is measured with the majority of detectors set to a first setting with respect to the spectral detection bands. The excitation wavelength is then shifted, and the resulting fluorescence is again measured with the majority of detectors set to the first setting. This may be repeated several times; in the specific embodiment shown, excitation occurs sequentially at a total of five excitation wavelengths, and detection occurs with five detectors. This sequence forms a first run of excitation steps. In this run, (five) initial sampling points of the excitation spectrum are obtained by determining the sum of the individual detector signals for each excitation wavelength.At the same time, an emission spectrum with several (five) support points is obtained for each of the (five) excitation wavelengths; according to the disclosure, information is obtained that could be used to determine the emission spectrum. According to a graphical representation of the relationship between the excitation wavelengths and the detection bands, in the first pass, the detection band of a first detector begins just above the first excitation wavelength, that of the second detector begins just above the second excitation wavelength, and so on. A second pass is then performed accordingly, whereby, on the one hand, the detection wavelength bands of all detectors are shifted by an amount, and, on the other hand, the excitation wavelengths are shifted by the same amount.The wavelength band of the longest-wavelength detector becomes narrower, as shown in the graphic, because the lower limit wavelength of the detection range is changed, but not the upper one. In this second pass, (five) additional support points of the excitation spectrum are obtained. Further passes can now be performed with further shifted excitation wavelengths and detection wavelength bands. The shifting is carried out until a further shift would result in the first excitation wavelength of the following pass corresponding to the second excitation wavelength of the first pass. This means that each individual excitation wavelength is used with exactly one setting of the spectral wavelength bands of the detectors. The passes can and should preferably be performed during scanning imaging of the sample.Preferably, the scanning and acquisition of the sample is carried out in such a way that a complete image of the sample is created precisely with the complete acquisition of the spectra. The document further states that an emission spectrum is obtained for each excitation wavelength from the signals of the individual detectors, in the aforementioned embodiment one with five support points. The number of support points can be increased by increasing the number of individual detectors. It is further claimed that dyes with identical emission spectra can be differentiated based on different excitation spectra, which makes it possible to image the dyes individually in a multiply dyed sample. A closer look, however, shows that the method specifically specified as the only embodiment is flawed and that the statements regarding application to multiply dyed samples are not applicable.This is because shifting the lower limit wavelength of the detection range with the shifting of the excitation wavelengths from run to run changes the detected wavelength range from run to run; it becomes smaller, so that the values ​​obtained in the second run will systematically and falsely be disproportionately small compared to the values ​​obtained in the first run. If multiple dyes are present in the sample, determining the excitation spectrum of a single dye is simply impossible, regardless of the problem described above. This is because multiple dyes are excited during excitation, and accordingly, the detectors always detect a combined signal. However, decomposing the combined signal into its individual summands is only possible if the excitation spectra of the individual dyes are known.If this were the case, the ratios in which the dyes are present in the excitation range could be determined. However, the actual procedure requires the excitation spectrum to be determined first.

[0004] Microscopes, particularly confocal laser scanning microscopes with detection units whose spectral sensitivity range is adjustable, are frequently mentioned in the state of the art. One example is German patent application DE 10 2006 034 908 A1. The detection wavelengths are usually adjusted depending on the fluorophores present in the respective sample. The adjustability of the filters then ensures that only components need to be adjusted before image acquisition of a sample, rather than replaced.

[0005] MINFLUX nanoscopy is a relatively new microscopy technique. MINFLUX nanoscopy is a localization microscopy technique. Fluorophores are localized using structured excitation light distributions. A fundamental feature of MINFLUX nanoscopy is that the fluorophores are excited in such a way that the fluorophore to be localized is always positioned close to or within a minimum of the excitation light distribution, which ideally is a zero point, with the excitation light distribution exhibiting a region of increasing intensity adjacent to the minimum. This allows for better utilization of the fluorescence photons to obtain information about the position of the respective emitting fluorophore. This also applies to applications in which the movement of fluorophores is to be tracked over time.The observation of a sample using an excitation minimum, the basis of MINFLUX nanoscopy, is known from patents DE 10 2011 055 367 B4, initially only for tracking the movement of individual molecules in a sample, and DE 10 2013 114 860 B3. DE 10 2011 055 367 B4 also addresses the possibility of tracking two different fluorophores simultaneously. For this purpose, two different light sources are used, which are quickly switched between.

[0006] Based on this foundation, a number of refinements for information acquisition have emerged that enable the localization of fluorophores with an uncertainty in the range below 2 nm. This uncertainty corresponds to the extent of fluorophores. A detailed description of MINFLUX nanoscopy can be found in "Nanometer resolution imaging and tracking of fluorescent molecules with minimal photon fluxes," Francisco Balzarotti et al., arXiv:1611.03401 [physics.optics] (2016). In principle, in order to localize a fluorophore using MINFLUX nanoscopy, the intensity minimum, or zero, must be placed at several positions relative to the fluorophore's position. For this purpose, a fluorophore position must be estimated with an initial, lower degree of accuracy in a preparatory step.This can be achieved, for example, using conventional localization microscopy (PALM, STORM) or other known methods. The cited publication describes a method in which a sample is scanned with a Gaussian intensity distribution until fluorescence is detected at a scanning position, which with a certain probability originates from an isolated molecule. The scanning is then stopped, and the intensity distribution is subsequently positioned at several locations—specifically, at four locations—around the scanning position at a distance of less than the wavelength of the light. From the photon numbers measured at the individual positions, the position of the isolated emitter is estimated using a method described in the publication. This estimation essentially corresponds to a ratiometric evaluation of the photon numbers.Subsequently, an intensity distribution of excitation light with a central minimum, for example, in the shape of a donut, as known from STED microscopy, is placed at a known position, which is chosen such that the fluorophore is close to the minimum of the intensity distribution. The fluorescence response of the fluorophore is measured. The same process is repeated for one or more additional positions of the intensity distribution. Using a ratiometric evaluation of the intensity ratios, the position of the fluorophore is determined with greater accuracy. In general, the emission rate increases the further the fluorophore is from the excitation minimum or the further the fluorophore is shifted into a range of increasing intensity.This more precisely determined position can now be used as the starting position for repeating the sequence of the aforementioned steps, whereby the majority of positions can be placed closer to the estimated fluorophore position. Particularly with regard to tracking the movement of fluorophores, the change in emission rate upon displacement of the fluorophore into the intensity increase region or toward the minimum can also be used to estimate the displacement of the fluorophore with high precision. The closer the minimum positions of the intensity distribution are to the actual location of the fluorophore, the fewer fluorescence photons are required for localization with a given uncertainty or accuracy. The MINFLUX method can also be applied for localization in three dimensions, for example, using so-called bottle beams for excitation.

[0007] Patent publications on MINFLUX nanoscopy include, in particular, WO 2018 / 069 283 A1, US 2019 / 0235220 A1, US 2019 / 0234882 A1, and US 2019 / 0234879 A1. These US patent applications are all follow-on applications to the first-mentioned provisional international patent application, in which all concepts mentioned below with reference to the US patent publications are also disclosed. All of these documents state that a STED microscope can be used to carry out the MINFLUX method, with the light that is fluorescence-preventing light when used as a STED microscope being used as the excitation light.

[0008] US 2019 / 0235220 A1 is directed to a method with a small or minimal number of positions at which an intensity minimum, which is adjacent to both sides of intensity increase regions in each spatial direction in which a position of the fluorophore is to be determined, is placed in order to determine the position of the fluorophore.

[0009] US 2019 / 0234882 A1 is directed to the method described above, in which the spatial information obtained from a first MINFLUX step is used to place the minimum of the intensity light distribution closer to the fluorophore in a subsequent step and to derive more precise spatial information from this.

[0010] US 2019 / 0234879 A1 addresses a method in which the intensity minimum is placed very quickly, almost simultaneously, at a number of positions around the estimated position of the fluorophore. A single position is then moved closer to the presumed minimum if an increased emission rate is detected at that position.

[0011] The publication "MINFLUX nanoscopy delivers 3D multicolor nanometer resolution in cells," Klaus C. Gwosch et al., Nat Methods 17, 217–224 (2020), https: / / doi.org / 10.1038 / s41592-019-0688-0 ,focuses on three-dimensional nanoscopy with several different fluorophores. The publication also documents evidence that iteratively approximating the minimum positions to the actual location of the fluorophore in question reduces the uncertainty of localization compared to ratiometric localization with fixed minimum positions. It is further demonstrated that this approach achieves isotropic resolution in three dimensions. For imaging using different fluorophores, dyes are used that exhibit a large overlap of their spectral absorption curves and can thus be excited with the same excitation wavelength.Specifically, three-dimensional nanoscopy is demonstrated with two different fluorophores each, using a total of two different pairs (A: CF660C and Alexa Fluor 647; B: CF680C and Alexa Fluor 647), with one partner of each pair being the dye Alexa Fluor 647. During MINFLUX imaging, the same laser is used to excite both dyes of a pair. The two dyes of each pair differ in their emission characteristics. The detection path of the nanoscopy arrangement used is split by a dichroic beam splitter into a spectral detection channel for light with a wavelength above 685 nm and one for light with a wavelength below 685 nm.During the measurement, light is detected in both spectral channels for all fluorophores used, since the spectral emission of all fluorophores used is significantly different from 0 (zero) both above and below the cutoff wavelength. The sum signal is used to localize the fluorophore. Since only one fluorophore usually fluoresces in the localization region, apart from background fluorescence, and since the different fluorophores of a pair differ in their spectral emission, the ratio of the signals in the different detection channels can be used to determine which fluorophore type of the dye pair caused the signal. As a result, a specific dye can be assigned to each localization. Potential misassignments are minimized during data analysis using a principal component analysis of the spectral components of all MINFLUX iterations.Since localization is based on knowledge of the positions of the minima of the intensity distributions of the excitation light alone, there is no color-dependent offset between the localizations of the different fluorophores, as they are excited with identical excitation wavelengths. The experimental setup features two different lasers for excitation with an excitation donut, one with a wavelength of 642 nm and one with a wavelength of 560 nm. It is demonstrated that MINFLUX nanoscopy can also be performed using photoactivatable fluorescent proteins, specifically mMaple, especially on living cells. The excitation laser with a wavelength of 560 nm is apparently required for the excitation of the fluorescent protein mMaple in the corresponding single-color MINFLUX image.

[0012] A largely identical procedure to that described in the above-mentioned publication is used to examine other samples in the publication "Multicolor 3D MINFLUX nanoscopy of mitochondrial MICOS proteins", Jasmin Pape et al., PNAS, 117 (34) 20607-20614, (2020), https: / / doi.orq / 10.1073 / pnas.2009364117 , To image densely packed structures, a special algorithm is used to form localization clusters or assign localizations to individual fluorophores. To select areas to be examined with MINFLUX, a wide-field image is first acquired. A camera and a separate laser are used for this purpose.

[0013] Using MINFLUX nanoscopy, the position of fluorophores in two spatial directions could be determined experimentally with an uncertainty of only 1 nm, meaning that the accuracy of the position determination is comparable to the extent of the fluorophores themselves. If the position of a single fluorophore is to be determined with a given measurement uncertainty, less time and, in particular, a smaller number of fluorescence photons are required than for the position determination of a single fluorophore using conventional localization microscopy.

[0014] European Patent Application EP 3 372 990 A1 describes, in addition to other MINFLUX methods, a method similar to the MINFLUX method in one aspect. As with a MINFLUX method, an isolated molecule is subjected to a focused intensity distribution having a central minimum, preferably a zero, and surrounding increasing regions. Also as with MINFLUX, this minimum is placed at a plurality of sampling points around the presumed location of the isolated molecule, and a fluorescence emission is detected for each sampling point. The actual position is estimated with high precision from the intensity values ​​or photon numbers thus obtained. In contrast to MINFLUX, the intensity distribution is a distribution of fluorescence-inhibiting light, in particular STED light. This is applied together with excitation light, whereby the intensity distribution of the excitation light does not have a central local minimum.While with MINFLUX the fluorescence emission is higher when the isolated molecule is further away from the central minimum of the intensity distribution, with this method it is the other way around.

[0015] The publication "Super-resolution microscopy demystified," by Lothar Schermelleh et al., Nature Cell Biology, VOL. 21, 72-84 (2019), https: / / doi.org / 10.1038 / s41556-018-0251-8, provides a current overview of super-resolution microscopy techniques. The authors also include techniques that do not transmit spatial frequency content above the maximum spatial frequency of confocal microscopy, but merely offer improved transmission of high spatial frequencies within the frequency spectrum determined by diffraction limitation. These techniques include, for example, structured illumination microscopy (SIM) and image scanning microscopy. The authors state that none of the known super-resolution techniques is suitable for all imaging purposes and that, therefore, complementary techniques should be used to address specific questions.Super-resolution microscopy offers scientists the opportunity to test and refine biological models; descriptive research, alongside hypothesis-driven research, is therefore becoming increasingly important if (patho)biological processes are to be understood. This publication also mentions correlative methods, in which images from different imaging techniques, for example, a light-optical, super-resolution image, are correlated with an electron microscopic image.

[0016] A further overview of current techniques in super-resolution microscopy and nanoscopy, including combined techniques such as a combination of STED microscopy with atomic force microscopy, can be found in the review article "Optical nanoscopy," A. Diaspro and P. Bianchini, Riv. Nuovo Cim. (2020). https: / / doi.org / 10.1007 / s40766-020-00008-1 .

[0017] In the publication "High-resolution deep view microscopy of cells and tissues", H. Schneckenburger et al., Quantum Electron. 50 2 (2020), https: / / doi.org / 10.1070 / QEL17204 , Various methods and devices for imaging thick biological samples are proposed. It is found that high resolution is usually only achieved using objectives with a high numerical aperture. These have a short working distance, making high-resolution imaging of thicker samples impossible. The publication now proposes illuminating a sample using a plurality of distributed apertures. This allows a type of focal point to be created at any depth within a sample. The light originating from the focal area can then be detected, and high-resolution images can be generated. This is achieved using both STED-type and MINFLUX-type illumination.

[0018] For example, from the publication "Lens-free on-chip microscopy based on dual-plane phase retrieval", Cheng Guo et al., Opt. Express 27, 35216-35229 (2019), https: / / doi.org / 10.1364 / OE.27.035216 , A known method is used in which laser light transmitted through and influenced by a sample is captured directly on a spatially resolving detector without the use of optical imaging elements. An image of the sample can be calculated from the signal. This method achieves an image resolution of approximately 1 µm. The method is fundamentally unsuitable for imaging fluorescent samples.

[0019] Patent publication DE 10 2006 047 816 A1 discloses a method and device for high-resolution optical scanning of a sample stained with multiple dyes. In this method, the sample is imaged using a STED method. According to the publication, imaging of the multiple dyes can be performed sequentially, among other ways, by first taking a STED image of one dye by completely scanning a sample area to be examined, and then imaging a different dye during a subsequent complete scan. In one embodiment, a common wavelength can be selected for the excitation of two dyes. The fluorescence of the two dyes can be separated by a wavelength-sensitive element (filter), and the fluorescence of each dye can be recorded with a separate detector.The described device can have a detection device with multiple detectors that can be switched off individually and / or in groups. The device, designed as a scanning microscope, can be adapted for use with dyes with specific spectral characteristics. A similar method for high-resolution optical scanning of a sample stained with at least two dyes as in the aforementioned document is described in DE 10 2006 011 556 A1. This method is characterized by the fact that the excitation spectra and de-excitation wavelengths of the two substances differ from each other. OBJECT OF THE INVENTION

[0020] The invention is based on the object of providing improved methods for the microscopic or nanoscopic imaging of a sample stained with multiple dyes. In particular, the methods should enable a nanoscopic image with up to molecular resolution to be placed in the spatial context of microscopic images or to perform nanoscopic tracking of individual molecules in the spatial context of a microscopic image. This should also be possible for living biological objects. SOLUTION

[0021] The object of the invention is achieved by methods having the features of independent claim 1 and by a microscope having the features of claim 15. Dependent claims 2 to 14 relate to preferred embodiments of the method according to claim 1. The present invention is defined in claims 1 and 15. DESCRIPTION OF THE INVENTION On the problem

[0022] Microscopy refers to techniques that allow objects to be imaged with high resolution. High resolution simply means that details of an imaged object that cannot be resolved with the naked eye can be resolved. If there is no visual observation, but data is recorded from which an image can be generated for viewing, the magnification of the image by the optical components of the microscope loses its significance with regard to image generation; the only things that matter are the optical resolution and the distance between the data points in the sample. In addition to the resolution of the microscope, the size of the field of view, or more generally the volume that can be imaged, is crucial for its applicability for the examination of a specific object or, in other words, for a specific examination task.When examining objects such as living cells or objects within living cells that change over time, the time required to capture an image becomes crucial. The requirements of a microscope therefore depend in many ways on the examination task. Accordingly, there are a multitude of different microscopes and microscopy techniques.

[0023] For a variety of microscopy techniques, resolution is limited by diffraction. In the field of fluorescence microscopy, the Rayleigh criterion is usually used to quantitatively describe the diffraction limit; in the context of the observation of scattered light, the Abbe diffraction limit is frequently mentioned. Although the resolution limit depends on the details of the observation conditions, it is always proportional to the wavelength of the light used and, in practically relevant applications, is approximately half the wavelength of the light used. The resolution of a microscope can also be described by specifying the optical transfer function (OTF) or the modulation transfer function (MTF), which is the absolute value of the OTF. The MTF has a cutoff frequency above which it constantly has the value 0. For fluorescence microscopy with wide-field illumination, this cutoff frequency is fn= 2 NA / λ, where NA is the numerical aperture of the imaging optics and λ is the wavelength of the fluorescent light; in the case of structured illumination, which basically also includes confocal illumination, the cutoff frequency is twice as large, i.e. fn= 4 NA / λ. The latter applies under the assumption that the excitation and emission wavelengths do not differ significantly. This cutoff frequency is absolute for diffraction-limited techniques and cannot be exceeded by special illumination structures, special imaging modalities, or deconvolution techniques, unless deconvolution is based on prior knowledge acquired elsewhere. In the case of simple confocal fluorescence microscopy, it should be noted that this cutoff frequency would only be achieved with an infinitesimally small pinhole, i.e., it cannot be achieved in practice. In practical applications, a pinhole is often used whose diameter corresponds to the diameter of the diffraction disk of the aperture stop up to the first zero. In this case, the cutoff frequency of the MTF in confocal fluorescence microscopy corresponds exactly to that in wide-field fluorescence microscopy. These statements refer to the lateral resolution.The same applies to the axial resolution, although the resolution in the axial direction is generally worse, i.e. the minimum distance between separable details is larger and the cutoff frequency of the modulation transfer function is lower.

[0024] This resolution limit can be exceeded using various known super-resolution techniques. These include STED microscopy, localization microscopy techniques such as PALM or STORM, and techniques using structured illumination and fluorescence saturation. However, the resolution of these techniques still depends on the wavelength and is limited by the available photon budget. In practice, such techniques can achieve resolutions in the range of 30 nm down to, in more specialized cases, 20 nm.

[0025] As explained in the prior art section, MINFLUX nanoscopy now enables imaging with a resolution in the sub-2 nm range. Such resolution is referred to in this application as molecular resolution because the 2 nm dimension roughly corresponds to the size of individual fluorescent dye molecules; many dyes have even more extended molecules. It is expected that molecular resolution can also be achieved with methods that employ MINFLUX-like sensing and localization using a distribution of fluorescence-inhibiting light with a central local minimum, as described, for example, in the above-mentioned document EP 3 372 990 A1. This method is referred to as STED-MINFLUX in this application.However, both methods are ill-suited for imaging typical biological samples in their entirety, primarily due to their time-consuming nature. Typically, only sections of such an examination can be examined using MINFLUX or STED-MINFLUX. If conventional localization microscopy techniques such as PALM or STORM are available with fluorophores that can emit a large number of photons before they are bleached, it is theoretically conceivable that molecular resolution could be achieved with such techniques. However, such techniques are also ill-suited for imaging typical biological samples in their entirety, particularly due to their time-consuming nature.In order to be able to image biological samples in their entirety and at the same time to image at least particularly interesting sections with molecular resolution, it is therefore necessary to combine other imaging or microscopy techniques with a method that achieves molecular resolution, in particular with MINFLUX, for example.

[0026] However, such a combination presents several problems. For example, it is desirable, and in certain cases necessary, given the objectives of the sample investigation, that the image information obtained with the other technique can be spatially contextualized with the molecularly resolved image information with great accuracy. However, this requires high-quality imaging of the sample with the other technique before imaging at molecular resolution. This creates the problem that imaging with the other technique places stress on the sample, meaning that the fluorophores it contains are bleached.There is generally a correlation between the degree of bleaching and the resolution achieved with the other method. This is because even with methods that are fundamentally diffraction-limited, the actual resolution achieved depends on the photon budget used. If only a few photons are used, the image quality is reduced by noise. This noise results, on the one hand, from the emission process itself, which results in Poisson-distributed noise in the measured photon numbers. On the other hand, however, it also results from the fact that detectors themselves exhibit noise, which has a greater influence on image quality the lower the photon numbers or intensities to be measured. At the same time, since imaging with molecular resolution requires that the fluorescent dyes used for this imaging are not or only slightly bleached, a conflict of objectives arises.

[0027] To image biological samples or the structures they contain, it is often necessary to stain the sample to be examined with several different dyes. This allows different structures to be clearly distinguished from one another in fluorescence microscopy images, while simultaneously visualizing their spatial relationship to one another. In order to be able to image such multiply stained samples at molecular resolution in a way that clearly distinguishes the different biological structures associated with different dyes, it is necessary to be able to distinguish between the fluorescence signals emitted by different dyes.Enabling this in a simple manner for a wide variety of samples is complicated by the fact that a device and method that can easily distinguish two specific dyes may be unsuitable for separating other dyes. A further problem is that even if the spectral properties of the dyes used to color a sample are known in principle, significant deviations in the spectral properties of the dyes can occur depending on the sample, e.g., the pH value of the sample. Therefore, a separation with predetermined device and method properties is not optimal.

[0028] The possibility of combining complementary techniques is currently often reserved for specialized laboratories. According to the above-mentioned article "Super-resolution microscopy demystified" (Lothar Schermelleh et al.), examinations of samples using super-resolution microscopy techniques often require validation with studies performed on different platforms.

[0029] A single platform that is suitable for commercialization to a broader user base and that, depending on the sample section, enables the imaging of biological samples with both a diffraction-limited resolution of approximately 400 nm to 250 nm and a super-resolution of approximately 50 nm to 30 nm, as well as molecular resolution, is not yet known. Likewise, no methods are known that are suitable for offering the capabilities mentioned above and, at the same time, can be implemented in a microscope as a single platform as described above. This means that no corresponding methods are currently known for which a single platform could be easily implemented. To solve the problem

[0030] This is where the invention comes in, providing methods that are particularly suitable for use in a common single platform, i.e., a microscope. Furthermore, it provides a microscope as such a single platform. The conflict of objectives mentioned above is thereby resolved as far as possible.

[0031] The purpose of the method is to make it particularly easy to provide a single platform, i.e., a microscope, with which the method can be carried out. The method involves first acquiring image data of the sample or a partial region of the sample, determining and selecting a sample region of interest based on the image data, and detecting fluorescence photons emitted from the sample in at least two different wavelength ranges. The detection is preferably performed in contiguous wavelength ranges and is not spectrally resolved for each wavelength range.

[0032] In the context of this specification (i.e., the method and the microscope), the term "non-spectrally resolved" means that the fluorescence detected within the respective contiguous or adjustable wavelength range is evaluated independently of the wavelength of the detected photons. Therefore, no spectral information is collected and used within the adjustable wavelength ranges; in other words, no spectrum is recorded. The detected fluorescence can be summarized to form a total fluorescence value (e.g., a fluorescence intensity or a counting rate per unit time obtained by photon counting). It is also possible to split the detected fluorescence according to a parameter other than wavelength, for example, according to the fluorescence lifetime.

[0033] Furthermore, the method aims to provide optimization for at least one of the imaging types, i.e., the initial acquisition of image data or a localization microscopy technique. Overall, the method aims to optimally utilize or utilize the dyes present in the sample for imaging a sample, with the imaging encompassing both the imaging of structures in the size range of a few tens of micrometers to millimeters as well as the imaging of subregions with molecular resolution. Tracking a particle of a sample is considered a special form of imaging.

[0034] Specifically, the task is solved by a method for a localization microscopic examination or for the localization microscopic examination of a sample stained with several dyes, which comprises the following steps: a first acquisition of image data of the sample or a partial region of the sample, a determination and selection of a sample region of interest based on the image data, an excitation of fluorophores, in particular at least one of the plurality of dyes, in a first partial region within the selected sample region of interest with excitation light of a first excitation wavelength and a first detection of the fluorescence photons emitted by the excited fluorophores in a first wavelength range with a first detection unit and a second detection of the fluorescence photons emitted by fluorophores excited with excitation light of the first excitation wavelength in a second wavelength range, which differs from the first wavelength range, with one or the first detection unit (in other words with the first detection unit or with a further detection unit).

[0035] Such a method is known from the state of the art, for example, from the publication "Multicolor 3D MINFLUX nanoscopy of mitochondrial MICOS proteins" (Jasmin Pape et al.) cited in the prior art section and the associated supplementary information. There, the first image is captured exclusively using a wide-field method with an EMCCD camera.

[0036] In the first method according to the invention, the first recording can in principle be carried out in the same way, but it preferably comprises a scanning recording of image data.

[0037] In the context of the present specification (i.e., the first to sixth methods and the microscope), the term "multi-dye fluorophores" refers to fluorescent molecules of several different species which differ, for example, in their molecular structure and thus, in particular, have different excitation or emission spectra or a different fluorescence lifetime.

[0038] Compared to the prior art, the first method according to the invention is characterized in that that the first detection unit is designed such that the wavelength range (in particular the first wavelength range) in which it (i.e. the first detection unit) detects fluorescence photons is adjustable and that the method further comprises: (in particular subsequently) determining wavelength ranges (in particular further wavelength ranges) for an optimized ratiometric separation of the plurality of dyes during a localization microscopic image recording of the sample in a plurality of spectral detection channels.

[0039] In the context of this specification (i.e., the method and the microscope), "ratiometric separation" of dyes is understood to mean a method in which a fluorophore (or the fluorescent light emitted by this fluorophore) can be assigned to one of the several dyes by determining a ratio of the number of fluorescence photons or fluorescence intensities detected in the two differing wavelength ranges. In this context, the term "ratiometric separation" in the context of this specification also includes the case in which no fluorescence signal is detected in one of the wavelength ranges.The corresponding ratio would then be equal to zero or "equal to infinity," depending on whether it is defined as the quotient between the number or intensity detected in a first wavelength range and the number or intensity detected in a second wavelength range, or whether it is defined as the reciprocal of the aforementioned quotient. Of course, in this case, too, the assignment of a fluorophore to a specific dye can be based on the ratio. Furthermore, the above-mentioned ratio does not necessarily have to be formed to achieve ratiometric separation in methods not according to the invention. The signals measured in the two different wavelength ranges can also be evaluated in other ways in such methods not according to the invention in order to assign a fluorophore to one of the several dyes.According to one embodiment, the first wavelength range and / or the second wavelength range for the optimized ratiometric separation of the dyes are adjusted or adapted on the basis of the determined (further) wavelength ranges.

[0040] The wavelength ranges determined according to the method for optimized ratiometric separation of the multiple dyes can then be used for a localization microscopy image, in particular a MINFLUX image, or they can alternatively or additionally be stored together with information about the sample, for example, in a database. In the sample region of interest, individual fluorophores (at least one of the multiple dyes) are localized or tracked using a MINFLUX method or a STED-MINFLUX method, with the fluorescence photons being detected in each of the wavelength ranges determined for optimized ratiometric separation of the multiple dyes.

[0041] In the context of the present specification (i.e., with regard to the first to sixth methods and the microscope), the term "MINFLUX method" refers to a method in which a local intensity minimum (in particular an intensity zero) of a light distribution (in particular excitation light distribution) is placed at a plurality of positions relative to the position of an individual fluorophore, whereby the position of the fluorophore is estimated from the photon numbers measured at the individual positions.

[0042] In the context of this specification (i.e., with regard to the first to sixth methods and the microscope), a STED-MINFLUX method is understood to mean a method in which a local intensity minimum of a STED (stimulated emission depletion) light distribution is placed at a plurality of positions relative to the position of an individual fluorophore illuminated with excitation light, whereby the position of the fluorophore is estimated from the photon numbers measured at the individual positions.

[0043] In the context of this specification, the localization of individual fluorophores refers to a method in which fluorescent fluorophores, i.e., fluorophores in an active state whose fluorescence photons are detectable in the excitation and emission wavelength range used, are spaced apart by a distance that corresponds at least to the optical diffraction limit at the wavelengths used. This requirement can be met, in particular, by selecting suitable dye concentrations for labeling the sample and, in the case of photoactivatable or flashing dyes, by selecting activation light parameters (e.g., power or pulse rate) and / or buffer conditions.The total density of fluorophores in the sample can be chosen such that certain fluorophores have a distance below the diffraction limit, as long as they are not active simultaneously or are excited and / or detected in different wavelength ranges.

[0044] Depending on the properties of the sample, such a method can be carried out in a first development in such a way that the first sub-region within the sample region of interest is selected on the basis of the image data such that it contains exclusively or substantially only fluorophores of a first dye, and that in the first sub-region a plurality of fluorophores are excited with excitation light of the first excitation wavelength, and that a first and a second value characterizing the spectral emission of the first dye are obtained from the first and second detections. Such a selection of the first sub-region is possible and useful if the first acquired image data depicts structures for which the user knows that only a single type of dye is to be expected.This allows the spectral properties of the dye in question to be determined very quickly in the specific sample environment, because this determination can then be carried out on an ensemble of numerous fluorophores of the dye in question. This provides a large and therefore low-noise signal without exposing the individual dye molecules of the dye in question to greater exposure to excitation light. Information can be obtained simply from the ratio of the number of fluorescence photons detected in the two differing wavelength ranges or the fluorescence intensities, which enables an optimized ratiometric separation of the different dyes. This applies both when the ratio of the intensities corresponds to a previously known intensity ratio and when it deviates significantly from such a previously known ratio.It is also possible that the measured values ​​obtained in the two wavelength ranges can be used to draw conclusions about background signals that affect the wavelength ranges to be defined for an optimized ratiometric separation.

[0045] This method according to the first development can now be advantageously carried out in one embodiment by additionally selecting a second sub-region and optionally further sub-regions within the sample region of interest on the basis of the image data in such a way that it or each of them contains exclusively or substantially only fluorophores of a second or a further dye and that in the second or each of a further sub-region a plurality of fluorophores is excited with excitation light of the first excitation wavelength and that a detection of the fluorescence photons emitted by the respective excited fluorophores takes place in the first and the second wavelength range and that in each case a first and a second value characterizing the spectral emission of the second or the respective further dye is obtained.In other words: Advantageously, as for the first dye, a range is selected for the second and optionally for further or all further dyes present in the sample in which exclusively or almost exclusively dye molecules of the second or the further dye in question are present. Even from the newly obtained information alone, conclusions of the same kind as for the first dye can be drawn. By comparing the ratio values ​​obtained for the various dyes, however, it is now possible to draw even better conclusions about the wavelength ranges to be defined for an optimized ratiometric separation. This embodiment can also be modified such that two new wavelength ranges are used instead of the first and second wavelength ranges.Excitation within each sub-area can preferably be achieved by scanning the sub-area or a section of the sub-area with focused excitation light. The use of focused excitation light facilitates selective excitation only within the sub-area. Scanning reduces local stress on the sample.

[0046] Both the first development and its configuration can now preferably be further developed in such a way that a further detection of the fluorescence photons emitted by fluorophores of the respectively selected dye, i.e. of the first, the second or a dye of the further dyes, each excited with excitation light of the first excitation wavelength, takes place in one or more further wavelength ranges which differ from one another and from the first and the second wavelength ranges, so that a plurality of values ​​characterizing the spectral emission of the first, the second or a dye of the further dyes is obtained from the measurement on the sample dyed with a plurality of dyes.In other words, advantageously, especially if the values ​​obtained from previous measurements still leave too much uncertainty regarding the wavelength ranges to be determined for optimized ratiometric separation, fluorescence measurements are performed for each of the dyes in at least one additional wavelength range. This improves the information about the emission spectra of the dyes in question, so that, in principle, a better determination of the wavelength ranges to be determined for optimized ratiometric separation is possible. However, it should be noted that each measurement through excitation and detection stresses the sample, i.e., it can bleach the sample, thus reducing the photon budget or the density of available fluorophores available for the subsequent measurement.The latter can be accepted if a different region within the sample, or even a different sample, can be used to determine the wavelength ranges to be defined for optimized ratiometric separation than the area ultimately to be examined by localization microscopy. It is also possible to use a sample that has already been satisfactorily imaged to determine the wavelength ranges to be defined for optimized ratiometric separation, so that the values ​​obtained from this can be used for other comparable samples.

[0047] In a second development of the method, the first sub-region within the sample region of interest is selected based on the image data such that it contains fluorophores of multiple dyes, and individual fluorophores are repeatedly excited in the first sub-region, whereby a total of fluorophores of all of the multiple dyes are excited. Excitation within the first sub-region can be carried out by scanning the sub-region or a section of the sub-region with focused excitation light. The use of focused excitation light makes it easier to excite selectively only within the sub-region. Scanning makes it easier to locate individual molecules that are in a state capable of fluorescence excitation during a particular period of time.Alternatively or additionally, activation light can also be used, for example, to illuminate the focal region, which is also exposed to excitation light, in order to switch fluorophores from a non-fluorescent state to an excitable state. Activation then does not occur as long as fluorescence can be detected from the individual molecule. Once this molecule has switched from a fluorescently excitable state to a non-excitable one, the activation can be repeated until fluorescence can be detected again from an individual molecule. Since multiple dyes are present in the partial region, fluorescence from the multiple dyes is detected upon multiple repetitions.

[0048] In principle, even in this second development, detection could be carried out in both the first and second wavelength ranges using the same detection unit. However, this appears to be difficult to implement and would place an increased strain on the sample. Therefore, particularly in this development of the first method, it is strongly preferred that the fluorescence photons in the first and second wavelength ranges be detected using separate (or separate) detection devices or detection units that are active simultaneously. This is also preferred in the first development. It has advantages over detection using just one detection device orA detection unit that is set once to the first wavelength range and once to the second wavelength range has the advantage that the acquisition of the spectral information is faster on the one hand, but on the other hand is also associated with less stress on the sample.

[0049] The second development of the method can now preferably be designed such that further excitation of fluorophores with excitation light of the first excitation wavelength and further detection takes place in two further wavelength ranges, wherein at least one of the two further wavelength ranges is neither the first nor the second wavelength range. This fundamentally improves the information that can be used to determine wavelength ranges for optimized ratiometric separation. Similar to a corresponding embodiment of the first development, however, it should also be noted here, for example, that each excitation leads to a load on the sample, which reduces the photon budget or the density of the available fluorophores that can be used for imaging with molecular resolution.

[0050] In both the first and second embodiments, it may be preferred that the first wavelength range and the second wavelength range are adjacent to one another at a first cutoff wavelength. It may then be further preferred that further wavelength ranges used for further detection are also adjacent to one another at a cutoff wavelength or a further cutoff wavelength, and it may also be preferred that further excitation and detection occur multiple times, with a different further cutoff wavelength being selected in each case.

[0051] Both developments can also advantageously be designed such that the first wavelength range and / or one of the further wavelength ranges comprises wavelengths that are shorter than the first excitation wavelength; preferably, they can be designed such that the first wavelength range and / or one of the further wavelength ranges exclusively comprises wavelengths that are shorter than the first excitation wavelength. In particular, with this preferred embodiment, comparatively little fluorescence can be detected in the relevant wavelength range. The advantage, however, is that, precisely in the wavelength range below the excitation wavelength, the fluorescence emission of different dyes can exhibit very large differences from one another, so that a ratiometric separation of the dyes is particularly well possible.For subsequent localization with molecular resolution, the other wavelength range will then essentially be used, while the wavelength range below the excitation wavelength will only be used to assign a localization to a dye. In this embodiment, the first wavelength range, if it exclusively comprises wavelengths that are shorter than the first excitation wavelength, will preferably not border on the second wavelength range. The same applies to further wavelength ranges used for further detection. Here, the wavelength range that exclusively comprises wavelengths that are shorter than the first excitation wavelength will not border on the other further wavelength range. Rather, the excitation wavelength will preferably not be contained in any of the wavelength ranges.This generally has the advantage that scattered excitation light cannot distort fluorescence detection. Furthermore, it allows the excitation light to be coupled into a beam path shared for excitation and detection using a color splitter, which can then preferably be designed as a narrowband color splitter with a bandwidth of a few nanometers, for example, less than 15 nm, 10 nm, or 5 nm.

[0052] The second development can preferably be designed such that fluorescence photons detected during detection in two wavelength ranges, in the first and second wavelength ranges and / or in one and another wavelength range of the two further wavelength ranges, are assigned to bursts of the isolated fluorophores, and that the numbers of fluorescence photons detected in one and the other wavelength range are determined for the bursts. Histograms are therefore formed. The wavelength ranges for an optimized ratiometric separation are then preferably determined from these histograms. It is known from the prior art how the quality of the separation of the dyes can be assessed from such histograms. If several histograms are available, it can then be assessed which histogram and thus which selection of wavelength ranges leads to the best ratiometric separation.

[0053] Advantageously, in all of the methods described above, the mutually different wavelength ranges do not overlap. This means that the first and second wavelength ranges do not overlap, and if repeated measurements are taken in multiple wavelength ranges, the wavelength ranges used in one repetition step do not overlap. This approach has the advantage of being easy to implement, for example, using a detection device with two or more individual detection units, as known from patent DE 10 2018 126 232 B3, cited at the beginning in the prior art section.

[0054] In the method and its two further developments, a recording step of first recording image data preferably takes place by scanning the sample or a section of the sample with an excitation light by means of a first deflection unit located in the optical path of the excitation light. The excitation light used in the first recording of image data can be the excitation light of the first excitation wavelength, but it can also be excitation light of a different wavelength, or it can have the first excitation wavelength but originate from a different light source than that of the excitation light of the first wavelength.The advantage of scanning using the first deflection unit, which is located in the beam path of the excitation light, is that the beam paths of the excitation light used for scanning during the first acquisition and the excitation light of the first wavelength can be more easily aligned with each other or maintained in an aligned state. This facilitates ensuring that the subregions selected in the image obtained during the first acquisition of image data are actually exposed to the first excitation light.

[0055] In a preferred embodiment, this applies expressly and mutatis mutandis to all methods according to the invention, but possibly as a mandatory feature, the first acquisition comprises a confocal scan or a joint scan with focused excitation and fluorescence-preventing light, preferably with STED light. Confocal scanning here means that the fluorescence emission is detected by confocal descanning. Confocal detection also occurs if a detection unit is arranged in the focal plane of the detection, which has a plurality of individual detector elements or apertures assigned to detectors, which are sized and arranged such that the diffraction pattern of a point light source is captured with spatial resolution. Such a detection unit can also be used in conjunction with the application of fluorescence-preventing light.Further preferably, detection during confocal scanning or joint scanning with focused excitation and fluorescence inhibition light occurs in a contiguous wavelength range and is not spectrally resolved. The advantage of this is that by selecting a contiguous wavelength range and foregoing spectral resolution within this wavelength range, the entire amount of light from the selected wavelength band is detected with exactly one detector, so that the signal obtained during the measurement process also contains the noise from only a single detector. This enables high imaging quality with low irradiation intensity. A further advantage arises when the sample is hardly known at the start of scanning. It is then not possible to select an imaging method that is already optimally tailored to the sample structure using multiple detection wavelengths.By selecting a contiguous wavelength range for detection, however, it can be ensured that all fluorescence emitted within that wavelength range is captured. Although this does not allow differentiation between different dyes, in the majority of cases an image dataset is obtained that is suitable for selecting regions of interest and subregions within a region of interest.

[0056] In a further embodiment, and this expressly applies to all methods according to the invention, the first acquisition of image data in the method comprises an acquisition step using a differential interference contrast method, preferably a scanning differential interference contrast method. A differential interference contrast method has the advantage that the fluorophores contained in the sample are not stressed, or at least only minimally stressed, during imaging, i.e., bleached. Selecting a scanning differential interference contrast method has the advantage that the first deflection unit can be used for the scanning process.This in turn allows, for example, the first acquisition of image data to include acquisition using a scanning differential interference contrast method, followed by the selection of a region of the sample, which can then be imaged as part of the first acquisition using confocal scanning or by combined scanning with focused excitation and fluorescence inhibition light, preferably with STED light. In the last image obtained, the region of interest and the subregion(s) are then selected, taking the differential interference contrast image into account. Since the same deflection unit is used for both methods applied during the first acquisition, the images are well aligned relative to each other.

[0057] Typically, detection is also performed descanned when imaging using a STED method, often using confocal detection. In addition to or as an alternative to confocal detection or descanned detection, wide-field detection can also be performed. If this is performed additionally, the image obtained with wide-field detection can be aligned with the image obtained with descanned detection.

[0058] In all methods according to the invention, the first acquisition of image data can also include an acquisition step using a wide-field fluorescence method, i.e., a method in which excitation also occurs using wide-field illumination. For example, an overview image can first be acquired using wide-field fluorescence microscopy, and a region can be selected in the overview image to be scanned confocally. During confocal scanning, wide-field detection can then additionally be performed, particularly in the edge regions of the selected region. The wide-field image and the scanned image can be aligned based on the image portions imaged both in wide-field and descanned mode. The region of interest and the sub-region(s) can then be selected in the aligned scanned image.

[0059] In a further embodiment, the method comprises localizing or tracking individual fluorophores using a MINFLUX method or a STED-MINFLUX method in the region of interest, with the fluorescence photons being detected in the wavelength ranges determined for optimized ratiometric separation of the multiple dyes. Alternatively or additionally, the wavelength ranges determined for optimized ratiometric separation of the multiple dyes are stored in a database, with the term "database" being interpreted broadly here. It is only important that the corresponding data are stored together with other data in such a way that they can be accessed for subsequent examinations of the same or other comparable samples.

[0060] It is also possible that the wavelength ranges to be used for optimized ratiometric separation of the multiple dyes change during imaging or that they differ in different sub-regions of the region of interest. According to the method, optimized wavelength ranges for ratiometric separation of the multiple dyes can be determined continuously and / or depending on the sub-regions to be imaged and applied accordingly. Such changes can result, for example, from changes in the fluorescence properties of the dyes used in the sample itself, but they can also result from changes in background fluorescence or, in particular, from a location-dependence of the frequencies of the fluorophores used. This can also depend on the specific, possibly location-dependent, measurement task.For example, in one area it may be important that a specific dye is detected particularly sensitively during localization, while at the same time it is not critical if the localization of another dye is wrongly assigned to the specific dye, and in another area it may be important that misassignments are excluded.

[0061] The use of multiple excitation wavelengths makes it possible to utilize multiple dyes with strongly different excitation spectra for multicolor image acquisition using a MINFLUX or STED-MINFLUX method. This is an important advantage, since one of the objectives of the invention is to place nanoscopic imaging with up to molecular resolution in the spatial context of microscopic images or to perform nanoscopic tracking of individual molecules in the spatial context of a microscopic image. For example, in confocal microscopy, fluorophore pairings with partners separated by excitation are regularly used. The method according to the invention also enables the use of such pairs for MINFLUX microscopy. Although the use of multiple excitation wavelengths instead of just one has a disadvantage with regard to the colocalizability of the different fluorophores, the advantage outweighs this disadvantage.At least the partial images recorded using one of the two excitation wavelengths used can each be well embedded in the image obtained during the first recording, i.e. with an accuracy corresponding to the spatial resolution of the first recording.

[0062] Finally, the microscope according to the invention is designed to carry out the method according to the invention. The microscope according to the invention has a laser configured to excite fluorescence, a scanning unit configured to scan a sample with focused excitation light, a detection unit configured to detect fluorescence photons, which are guided from the sample via an optical path into a plane conjugate to the focus of the excitation light, in a non-spectral manner within adjustable spectral ranges, and a device, preferably a variably adjustable wavefront modulator, configured to influence a wavefront of excitation light in such a way that the excitation light forms an intensity distribution with a central minimum at the location of the sample in a focus area, and a control unit configured to carry out a MINFLUX method.

[0063] The control unit is designed to carry out the method described above. Preferably, the microscope has at least two detection units, each configured to detect fluorescence within adjustable spectral ranges without spectral resolution, wherein preferably at least one of the detection units is configured to capture a diffraction image of a point object with spatial resolution.

[0064] The microscope preferably has a stand with a sample carrier, an objective lens held in an objective changing device, and a tube lens, wherein the scanning unit, the tube lens, and the objective lens are arranged along an optical path, and wherein the detection unit is arranged in an optical path that includes the objective lens, the tube lens, and the scanning unit. This design has the advantage that basic components that can also be used for microscopes not according to the invention can be used to a large extent for the microscope according to the invention. The microscope preferably has a plurality of different objectives in the objective changing device. This makes it possible to use objectives of different magnifications in order to capture image fields of different sizes.For example, the first recording in the process can easily be carried out in such a way that first an overview image is taken with a lens of low magnification and then a detail image is taken with a lens of higher magnification.

[0065] The use of a stand also has the advantage that additional components can be added relatively easily. For example, a detector for capturing images while detecting a differential interference contrast can be attached to the stand. Accordingly, the microscope preferably has such a detector for detecting a differential interference contrast, which is detachably attachable to the stand. It should be detachably attachable so that when not needed, it does not take up space in an optical path and prevent its use for other purposes, such as direct visual observation by an observer. If the microscope has such a detector, the control unit of the microscope is preferably connected or connectable to the detector for detecting a differential interference contrast and is also configured to control the detector and process its detection signals.

[0066] In one variant, the microscope features a wide-field detector. When using a stand, this can be connected to a camera port on the stand, for example.

[0067] Preferably, the microscope has a further deflection unit arranged in an optical path for illuminating the sample and not in the path to the detection unit. This unit is configured to enable extremely fast scanning, particularly useful for carrying out the MINFLUX or STED-MINFLUX method, but with very small areas on the order of 1 µm. Such a unit can, for example, be an electro-optical deflection unit or be constructed from several electro-optical deflection units.

[0068] Alternatively, the first deflection unit can also be configured to perform the small-scale scanning. While this generally slows down the scanning, it significantly reduces the equipment required. This variant can be particularly useful since the microscope according to the invention is also designed to acquire high-resolution or, particularly with the integration of a STED laser, super-resolution images using methods other than the MINFLUX method. This variant can be useful because the MINFLUX method is not necessarily the primary application and is only used when specifically needed, thus reducing the time loss.

[0069] Preferably, the microscope then has a pulsed laser for illuminating the sample with fluorescence-preventing light, preferably STED light. Further preferably, the control unit is configured to perform a STED-MINFLUX method.

[0070] From the above explanations of preferred embodiments of the method according to the invention, corresponding preferred embodiments of the microscope according to the invention result. DESCRIPTION OF THE FIGURES AND EXPLANATION OF THE INVENTION WITH THE FIGURES

[0071] In the following, the invention is further explained and described with reference to preferred embodiments shown in the figures. Fig. 1 shows schematically a microscope according to the invention Fig. 2 schematically illustrates the process of imaging a sample starting from a large image field to imaging a section with molecular resolution Fig. 3 illustrates embodiments of a fifth, fourth and third method Fig. 4 illustrates embodiments of a second and a first method

[0072] The Fig. 1The microscope 1 according to the invention shown has a confocally arranged detection unit 2, which has an adjustable wavelength range and is configured to detect light of a contiguous wavelength range at each selected setting. It further has a light source unit 3. This light source unit 3 comprises an excitation laser and a unit for modulating the wavefront of the excitation light 31. The excitation light 31 is guided along an illumination light path 21 onto a dichroic element 15. By means of the dichroic element, the excitation light 31 is coupled into a section of a light path that forms a common light path 23 for the excitation light 31 and for the detection light 32.Located in the common light path 23 is a deflection unit 4, which is designed to scan the excitation light 31 across a larger image field during confocal microscopy imaging, as well as to perform small scanning movements for performing a MINFLUX method. The excitation light 31 is focused onto a sample 7 located on a sample carrier 6 via a deflection mirror 11, a tube lens 12, and an objective 5, which is mounted in an objective changing device 13. If the excitation light 31 is suitable for exciting fluorescence in the sample 7, the fluorescence excited in the sample 7 is guided as detection light 32 along the same common light path 23 in the opposite direction to the dichroic element 15.The detection light 32 passes through the dichroic element 15, while any scattered excitation light 31 is prevented from passing through and is deflected toward the light source unit 3. The detection light 32 is guided along the detection light path 22 to the confocally arranged detection unit 2 and detected by it. The microscope further comprises a control unit 14, which is configured to carry out a MINFLUX method and to carry out a method according to claim 1. At the same time, it is configured to carry out confocal image acquisition. It is also connected to the automatically controllable components of the microscope 1, such as the deflection unit, as well as to the detection unit 2 and to a further detector 9.Connected here means that signals can be sent between the control unit 14 and the connected components and detectors in at least one direction, for example, via radio or a cable connection. The microscope 1 further comprises a stand 10, which in particular carries or contains the tube lens 12, the objective 5, and the sample carrier 6. The further detector 9 is detachably attached to the stand 10. This detector 9 detects light 34 transmitted through the sample 7, which is guided onto it along a transmitted light detection light path 24 through a condenser 8.

[0073] Fig. 2schematically illustrates the sequence of a method according to the invention. The first acquisition 100 comprises, in the example shown, three sub-steps 101, 102, 13. A sample 7, which here completely fills an entire area 70, is imaged in a sub-step 101 of the first acquisition in order to find and, if necessary, identify objects to be examined. In sub-step 101, a scanning contrast method can be used. For the illumination, an objective 5 with low magnification, for example 10x, is selected, which ensures that the scanning unit 4 can completely scan the entire area 70. The detection can be carried out with the Figure 1illustrated detector 9. By utilizing optical components (not shown) in the optical path 21, 23 in the illumination direction upstream of the scanning unit 4, i.e., in such a way that the illumination light first reaches the components and then the scanning unit 4, and in the transmitted light detection light path 24, differential interference contrast can be applied as a contrast method. Before switching to a fluorescence microscopic imaging method, these components (not shown) are removed at least from the detection light path 22, including the common light path 23, for example, pivoted out by a motor. In the overall area 70, an object is found in a partial area 71. This partial area 71 is then scanned and imaged in a second sub-step 102 using an objective 5 with higher magnification, for example 100x.For this purpose, the same or a different imaging method can be used than in the first step. Preferably, a method is selected that places as little stress on the sample 7 as possible. In the partial region 71, there is a cell with a cell membrane 711, organelles 712 (thin lines), further organelles (thicker lines, without reference numbers for clarity), and a cell nucleus 713. The observer now selects a partial region 72 within the cell that contains organelles 712. This partial region is then imaged using a high-resolution scanning method, for example, using confocal microscopy or STED imaging. A sample region of interest 80 is identified in the resulting image.This sample region 80 of interest is chosen here such that within it both the measurements for determining wavelength ranges for an optimized ratiometric separation of the multiple dyes are carried out and the localization 200 of the fluorophores 90 themselves.

[0074] In Figure 3Flow diagrams of a fifth method 45, a fourth method 44 and a third method 43 are shown. In addition, a decision tree is shown above, from which it can be seen under which conditions the respective method is preferentially used. The left-hand method is used in particular when the microscope does not have parallel, suitable detection channels, but when it only has exactly one suitable detection unit. If the spectral properties of the two dyes contained in sample 7 are basically known and the absorption maxima are far enough apart to separate the dyes by choosing different excitation wavelengths, for example more than 120 nm, then the fifth method 45 can advantageously be used. In the flow diagram, steps to be carried out in chronological sequence are marked by a directed arrow.If steps within a box are shown one below the other but not connected, the order is not fixed. An exception to this is the information for the fifth method 45 regarding the selection and adjustment of the excitation lasers and the performance of the MINFLUX localization. Here, it is clear that MINFLUX localizations of dye A are performed using settings for dye A, and the same applies to dye B. The flowchart is self-explanatory in light of this information.

[0075] If, however, microscope 1 has suitable parallel detection channels, and the spectral properties of the two dyes contained in sample 7 are known in principle, and the absorption maxima are sufficiently far apart for separation of the dyes by selecting different excitation wavelengths, for example, more than 120 nm, then the fourth method 44 can be advantageously applied. The flow diagram is self-explanatory.

[0076] In the event that wavelength ranges for optimized ratiometric separation of the multiple dyes are already very well known in advance, for example, because they were previously determined on a comparison sample using a method according to the invention, one problem to be solved is to perform an overall imaging that is as photon-efficient as possible, which in the image provides a highly precise embedding of the region acquired using a MINFLUX method into the image region obtained during the first acquisition 100. In particular, in the event that the absorption spectra of the multiple dyes used are too similar for separation using different excitation wavelengths, this problem is solved by a third method. A prescan with broadband detection is performed in a first detection channel, and a sample region 80 of interest is further determined.Typically, the user will first determine the sample region of interest 80 before setting the two detection channels to be used subsequently. Therefore, the corresponding steps are connected here by a directed arrow. In fact, the order can also be reversed. After the first detection channel has been changed and the second detection channel is set, a MINFLUX localization is performed using two detection channels. The fluorophores 90 are separated ratiometrically. It is also conceivable to perform detection during the prescan with a detection unit that is not used for MINFLUX localization. However, this not only requires an additional detection unit but also increases the adjustment effort.If detection is performed with a cascaded detection device, in particular one such as that known from patent DE 10 2018 126 232 B3, detection during the pre-scan is preferably performed with the first detection unit. Further detectors are only reached by light with wavelengths that are not directed to this first detection unit. Consequently, for detection in two channels, which often overlap at least with the first wide detection channel, the first detection channel must be changed.

[0077] In Figure 4Flow diagrams of embodiments of a second method 42 and a first method 41 are shown. These flow diagrams are also largely self-explanatory against the background of the corresponding explanations of the respective method 41 in the description of the invention. However, it should be explained here what is indicated by the two dashed arrows and the line connected to them within the frame symbolizing the second method 42. Between different steps of localization in several detection channels, the detection channels are changed. On the one hand, this can be done in order to find optimized settings by comparing the quality of the ratiometric separation. However, it can also be done if it is recognized from the data that the settings must be changed for an optimized separation.In this case, optimized or at least improved detection channels are first determined before the settings are changed. It should also be noted that the first localization, which is simultaneously used to find improved or optimized detection channels, can also be a method such as PALM. In principle, this also applies to subsequent localizations. However, such methods can only achieve molecular resolution under very favorable conditions. Therefore, the invention proposes the use of a MINFLUX or a STED-MINFLUX method. LIST OF REFERENCE SYMBOLS

[0078] 1Microscope 2Detection device or detection unit 3Light source unit 4Scanning unit 5Objective 6Sample carrier 7Sample 8Condenser 9Detector 10Stand 11Deflection mirror 12Tube lens 13Objective changing device 14Control unit 15Dichroic element 21Illumination light path 22Detection light path 23Common light path 24Transmitted light detection light path 31Excitation light 32Detection light 34Transmitted light 41First method 42Second method 43Third method 44Fourth method 45Fifth method 70Total area 71Partial area 72Partial area 80Specimen area of ​​interest 90Fluorophores 100First acquisition 101Substep 102Substep 103Substep 200Localize 411First step 412Second step 711Cell membrane 712Organelle 713Cell nucleus

Claims

1. Method for localization microscopic examination (200) of a sample (7) stained with multiple dyes comprising: - a first acquisition (100) of image data of the sample (7) or a partial area (101,102,103) of the sample, - a determining and selecting a sample region of interest (80) based on the image data, - an exciting of fluorophores (90) in a first subregion within the selected sample region of interest (80) with excitation light (31) of a first excitation wavelength, and a first detecting of the fluorescence photons emitted by the excited fluorophores (90) in a first wavelength region with a first detection unit (2), wherein a localizing (200) or a tracking of isolated fluorophores (90) is performed in the sample region of interest (80) according to a MINFLUX method or a STED-MINFLUX method, - a second detecting of the fluorescence photons emitted by fluorophores (90) excited with excitation light (31) of the first excitation wavelength in a second wavelength range different from the first wavelength range with a detection unit (2) or the detection unit (2), characterized in that the first detection unit (2) is configured such that the wavelength range in which it detects fluorescence photons is adjustable, and in that the method further comprises: - a determining of wavelength ranges for an optimized ratiometric separation of the multiple dyes in a localization microscopic image acquisition of the sample (7) in multiple spectral detection channels, wherein the detecting of the fluorescence photons is performed respectively in the wavelength ranges determined for optimized ratiometric separation of the plurality of dyes, wherein, in the ratiometric separation, a fluorophore is assigned to one of the plurality of dyes by determining a ratio of the numbers of fluorescence photons detected in the wavelength ranges.

2. Method according to claim 1, characterized in that the first subregion within the sample region of interest (80) is selected based on the image data such that only or substantially only fluorophores (90) of a first dye are contained therein, and in that a plurality of fluorophores (90) in the first subregion are excited with excitation light (31) of the first excitation wavelength, and in that a first and a second value characterizing the spectral emission of the first dye are obtained from the first and second detections.

3. Method according to claim 2, characterized in that a second subregion and optionally further subregions within the sample region of interest (80) is or are selected based on the image data in such a way that in it or in each of them exclusively or substantially only fluorophores (90) of a second or a further dye are contained and that in the second or respectively in a further subregion a plurality of fluorophores (90) is excited with excitation light (31) of the first excitation wavelength and in that detection of the fluorescence photons emitted by the respective excited fluorophores (90) takes place in the first and the second wavelength range, and in that in each case a first and a second value characterizing the spectral emission of the second or of the respective further dye are obtained.

4. Method according to claim 2 or 3, characterized in that a further detecting of the fluorescence photons emitted by fluorophores (90) of the first, of the second or of one dye of the further dyes respectively excited with excitation light (31) of the first excitation wavelength takes place in one or more further wavelength ranges, different from each other and from the first and second wavelength ranges, so that a plurality of values characterizing the spectral emission of the first, the second or one of the further dyes is obtained from the measurement on the sample (7) stained with multiple dyes.

5. Method according to one of the claims 2 to 4, characterized in that the alternately different wavelength ranges do not overlap.

6. The method according to one of the claims 3 to 5, characterized in that the wavelength ranges for an optimized ratiometric separation are obtained from the respective first and second values characterizing the spectral emission of the first or the second or optionally of a further dye or from the plurality of such values.

7. Method according to one of the claims 2 to 6, characterized in that the first detection unit (2) is used for detection in different wavelength ranges.

8. Method according to claim 1, characterized in that the first subregion within the sample region of interest (80) is selected based on the image data to include fluorophores (90) of multiple dyes, and in that isolated fluorophores (90) are repeatedly excited in the first subregion, wherein in total fluorophores (90) of all of the multiple dyes are excited.

9. Method according to claim 8, characterized in that a further exciting of fluorophores (90) with excitation light (31) of the first excitation wavelength and a further detecting in two further wavelength ranges takes place, wherein at least one of the two further wavelength ranges is neither the first nor the second wavelength range.

10. Method according to claim 8 or 9, characterized in that the first wavelength range and the second wavelength range are adjacent to each other at a first threshold wavelength.

11. Method according to claim 9 or 10, characterized in that the two further wavelength ranges are adjacent to each other at one or the threshold wavelength or a further threshold wavelength, in particular wherein a further exciting and detecting is carried out several times, with a different further threshold wavelength being selected in each case.

12. Method according to claim 8 or 9, characterized in that the first wavelength range and / or one of the further wavelength ranges comprises wavelengths that are smaller than the first excitation wavelength, preferably exclusively comprises wavelengths that are smaller than the first excitation wavelength.

13. Method according to one of the claims 8 to 12, characterized in that an assignment of fluorescence photons detected in each case during the detection in two wavelength ranges, in the first and the second wavelength range and / or in one and another wavelength range of the two further wavelength ranges, to bursts of the isolated fluorophores (90) is carried out, and in that the numbers of the fluorescence photons detected in the one and the other wavelength range are determined for the bursts in each case.

14. Method according to claim 13, characterized in that the wavelength ranges for optimized ratiometric separation are determined from the numbers of fluorescence photons detected in the one and the other wavelength range determined for the bursts.

15. Microscope (1) set up to perform a method according to one of claims 1 to 14 comprising: - a laser set up to excite fluorescence, - a scanning unit (4) set up to scan a sample (7) in a rasterized manner with focused excitation light (31), - a detection unit (2), which is configured to detect fluorescence photons, which are guided from the sample (7) via an optical path (22) into a plane conjugate to the focus of the excitation light (31), within adjustable wavelength ranges in a non-spectrally resolved manner, and - a device set up to influence a wavefront of excitation light (31) in such a way that the excitation light (31) forms an intensity distribution with a central minimum at the location of the sample (7) in a focal area, and - a control unit (14) set up to perform a MINFLUX method, wherein the control unit (14) is further configured to perform a method according to one of claims 1 to 14.