DE102005034443A1 - Sample e.g. cell particle, luminescence microscopy method, involves prevailing one of sample regions for image of sample, so that image has local resolution which is enhanced in relation to excitation radiation distribution - Google Patents

Abstract

The method involves displacing two partial regions of a sample in two states, respectively, while irradiation of excitation radiation (A) takes place with an excitation radiation distribution. An image of a luminescent sample has two sample regions in the states, respectively. One of the sample regions prevails for the image of the sample, so that the image has a local resolution which is enhanced in relation to the distribution. An independent claim is also included for a microscope for resolution enhanced luminescence microscopy of a sample.

Description

The This invention relates to resolution enhanced luminescence microscopy and, in particular, a procedure in which one to be investigated luminescent sample is illuminated with excitation radiation and an image of the luminescence excited sample is obtained. The The invention further relates to a microscope for resolution-enhanced Luminescence microscopy of a sample, the means for the excitation of Luminescence radiating excitation radiation in the sample and means for obtaining an image of the excited sample.

One classical field of application of light microscopy for examination of biological preparations is the luminescence microscopy. Here are certain dyes (so-called phosphors or fluorophores) for the specific labeling of Samples, e.g. of cell parts, used. The sample is, as mentioned, with Excited radiation illuminates and thereby stimulated luminescence detected with suitable detectors. Usually For this purpose, a dichroic beam splitter is combined in a light microscope provided with block filters which control the fluorescence radiation from the Separate excitation radiation and allow a separate observation. Through this procedure, the representation of individual, different colored Cell parts in the light microscope possible. Naturally can also several parts of a preparation simultaneously with different, specific to different Structures of the drug colored coloring matter become. This process is called multiple luminescence. Also, you can measure samples per se, so without dye addition luminesce.

Luminescence is here, as is common practice, as Generic term for Phosphorescence and fluorescence understood, thus capturing both processes.

Further It is known for sample examination, laser scanning microscopes (also abbreviated LSM) to use that from a three-dimensional illuminated image by means of a confocal detection arrangement (then one speaks of a confocal LSM) or a nonlinear sample interaction (so-called multiphoton microscopy) only reproduce that plane which is located in the focal plane of the lens. It will be a gained optical section, and recording multiple optical Cuts in different depths of the sample then allows, with Help of a suitable data processing device a three-dimensional image to generate the sample from the different optical sections is composed. Laser scanning microscopy is thus for Examination of thick preparations suitable.

Of course it will also uses a combination of luminescence microscopy and laser scanning microscopy, in which a luminescent sample in different depth levels is mapped with the help of an LSM.

in principle is the optical resolution a light microscope, even an LSM, by the physical Laws diffraction limited. For optimum resolution within these limits For example, special lighting configurations are known 4Pi arrangement or arrangements with standing wave fields. So that can the resolution, especially in the axial direction compared to a classic LSM clearly be improved. With the help of non-linear depopulation processes can continue the resolution to a factor of up to 10 versus a diffraction-limited one Confocal LSM be raised.

The 1a / b show such a procedure, as eg in US 5866911 is described. In this case, work is done to increase the resolution with a two wavelengths having light radiation. The light radiation of one wavelength is called the excitation light beam 1 focused on the sample to be measured by means of an objective and there stimulates luminescence, here fluorescence. The representation in the 1a / b shows for simplification only the one-dimensional case. The increase of the spatial resolution now takes place in that a light beam 2 at the other wavelength in sub-areas depopulates the excited by the excitation light fluorescent state. Therefore, this ray of light is also called "depopulation radiation." Now, the irradiation takes place, for example, in such a way that the main maximum of the depopulation light beam 2 and the main maximum of the excitation light beam 1 partially cover how it is in 1a easy to recognize. By this "de-excitation" of the sample at the edges of the excitation radiation 1 illuminated area, only sends a reduced volume 3 Fluorescence off, as in 1b good to see. The resolution is thus increased by this volume reduction.

2a -C show three possible mechanisms by which such depopulation can occur. In 2a the process of stimulated emission stimulation (STED) is shown. The excitation radiation is in the fluorophore of the level S1 (arrow A). The depopulation of the thus excited level S1 is accomplished in the direction of the basic level S0 by light radiation having a wavelength in the range of the fluorescence wavelength. Arrow SE shows this stimulated emission whose wavelength corresponds to that of luminescence (arrow F) almost identical. Thus, the excitation wavelength is around the amount of Stokesshiftes short-wave than that of the depopulation radiation. The resolution increase according to this approach thus requires two different light sources, as well as the prior art in the form of DE 4416558 C2 busy.

2 B illustrates another possible process of depopulation for the excited level S1 (arrow A) by exciting it to an even higher level S2 (arrow A +) which can no longer emit luminescence. This lifting is referred to in English as Excited State Absorption, which is why this procedure is also evidenced by the abbreviation ESA. A corresponding description of this process can be found, for example, in US 6633432 , As the distance of the energy levels in a sample or dye decreases to higher levels, for depopulation in the ESA process, a light source of lower energy and hence longer wavelength is used than for excitation. So you need two different light sources.

Another method for depopulation represents for fluorescence the so-called reversible saturable optical fluorescence transition, which, for example, in DE 10325460 A1 described and in 2c is illustrated. This approach uses a high resolution imaging dye with a switching beam 4 repeated from a first state 5 , in which fluorescence takes place, in a second state 6 in which the dye is not fluorescent is convertible, wherein the dye from the second state 6 in the first state 5 can return, like 2c illustrated. The sample with the dye is in partial areas with the switching beam 4 in the second state 6 transferred, wherein a defined region of the sample is omitted. With an excitation beam 1 then becomes fluorescent light 7 stimulated and subsequently registered. The fluorescent light 7 then comes only from sample volumes, not previously with the switching beam 7 were charged. By suitable overlap of excitation beam 1 and switching beam 4 is the volume from the fluorescent light 7 is emitted smaller than the resolution of the excitation beam 1 and the sharpness of the zero of the switching beam 4a allowed priori.

In all three mentioned procedures according to 2a to 2c Thus, the prevention of fluorescence by the use of light radiation with a wavelength that is unlike the wavelength for excitation. At the same time, this light radiation must have at least one well-defined local zero point of the radiation power, which determines the final resolution of the detected fluorescence radiation. As soon as the zero point is formed only as a minimum and does not completely return to zero, the fluorescence power and thus the efficiency of the method are further reduced. This is for example in the case of aberrations of the optical arrangement or in the preparation.

3 shows a known device using one of the three mentioned methods for increasing resolution, in the example of 3 the STED process. An excitation beam source 8th creates an Airy distribution in the sample 10 with which the sample is transferred from the basic level S0 to the excited state S1. The depopulation of the state S1 takes place by means of a depopulation light source 11 that under the use of a phase plate 12 a donut or torus-shaped beam distribution 13 in the sample 10 Has. The luminescence radiation of the non-depopulated, ie non-depleted, dye molecules is detected by means of a detector 14 detected. The depopulation increases the resolution of the microscope beyond the diffraction limit resulting from the Airy distribution. This comes through a reduced point spread distribution 15 of the high-resolution microscope compared to the conventional microscope 16 to expression.

Another method for increasing the resolution is in the EP 1157297 B1 addressed. In the process, nonlinear processes should be exploited by means of structured illumination. As a non-linearity, the document mentions the saturation of the fluorescence. The described method makes use of structured illumination to realize a displacement of the object space spectrum relative to the transfer function of the optical system. Specifically, the shift of the spectrum means that object space frequencies V0 are transmitted at a spatial frequency V0-Vm, where Vm is the frequency of the structured illumination. Given a maximum transferable spatial frequency by the system, this enables the transfer of spatial frequencies of the object lying around the shift frequency Vm above the maximum frequency of the transfer function. This approach requires a reconstruction algorithm for image generation and the exploitation of multiple images for an image.

Of the The invention is therefore based on the object, a Lumineszenzmikroskopieverfahren or to provide a luminescence microscope, which increases the resolution without recourse to several wavelengths or achieved without complex image reconstruction algorithms.

This object is achieved with a resolution-enhanced luminescence microscopy method in which a sample by irradiation of excitation radiation for the emission of certain luminescent radiation is excited and an image of the luminescent sample is obtained, wherein the luminescent sample from a first luminescence state in which the excitability for emission of the determined luminescence radiation increases with increasing excitation radiation power up to a maximum value, which is associated with an excitation radiation power threshold , can be converted into a second luminescence state, in which the sample has reduced excitability relative to the first state for emission of the determined luminescence radiation, wherein the sample can be brought into the second state by irradiation of excitation radiation power above the threshold value, the sample in partial regions is placed in the first state and in adjacent sub-areas in the second state by the irradiation of excitation radiation with an excitation radiation distribution is at least a local maximum power above the threshold and at least one local, local power minimum below the threshold, the image of the luminescent sample sample areas in the first state and sample areas in the second state, wherein the image of the luminescent sample predominantly contribute sample areas in the first state and thereby the image has increased spatial resolution compared to the excitation radiation distribution.

The Task will be solved further with a microscope for resolution-enhanced Luminescence microscopy, the means for excitation of luminescence, the excitation radiation irradiate the sample and thus the emission of certain luminescent radiation in the sample, means for obtaining an image of the luminescent sample, wherein the means for exciting the excitation radiation with a certain Radiate excitation radiation distribution that at least one local Achievement maximum that over a threshold, and at least one local, local minimum power which is below the threshold, wherein the threshold two luminescence regions of the sample separates, a first, at excitation radiation powers below the threshold state-state, in which the Excitability for the emission of certain luminescence radiation with increasing excitation radiation power up to a maximum value, which is reached at the threshold increases, and a second at and / or after excitation radiation powers above the threshold value present state region in which the sample one over the first Region reduced excitability for the emission of certain luminescence radiation and wherein the image acquisition means comprises sample areas in the first region, with excitation radiation below of the threshold were irradiated, and sample areas in the second region, the irradiated with excitation radiation power above the threshold value were to capture, taking the image of the sample predominantly sample areas in the first Contribute to the region and thereby increase the image compared to the excitation radiation distribution spatial resolution Has.

It So if a sample is used or the microscope is for a corresponding Sample designed, the luminescent material is substantially in two states can be located. In a first state, which occurs during the irradiation of Set excitation radiation power below the threshold and that is achieved by means of excitation, that radiates Material luminescent radiation, whose performance is usually with the excitation radiation power increases. In a second state, upon irradiation of excitation radiation power above the Threshold is present, either none or diminished Luminescence. Also can across from changed the first state Absorption properties and / or luminescence radiation emission with different optical properties than in the first state, for example with different spectral composition, polarization or lifetime, respectively. Relative to the sample, which naturally consists of a variety of composed of fluorescent or auto-fluorescent molecules, there are two status regions. In a first state region is the majority of molecules in the first state, in a second state region, the majority in the second state. According to the invention now sample areas into the first state region (or state for short) and other sample regions spent in the second state region (or state for short). The local Near the Areas that have been brought into the first state, ie on the Excitation radiation power falls below the threshold, and Regions in the second state, i. Areas with excitation radiation power irradiated above the threshold allows a significant Increase the resolution across from the excitation radiation distribution.

The achieved inventive solution So the increase in resolution by a nonlinear luminescent characteristic that emitted the Luminescence power as a function of excitation radiation power describes and has a local maximum. It is advantageous but not mandatory, if the characteristic curve on both sides This peak has relatively steep edges. Also can for suggestion Multiphoton process to Use come.

Since in the inventive approach as possible only sample areas in the first state contribute to the image luminescence and luminesce sample areas in the second state, at least to a lesser extent, is an on solution increase achieved because the sample image shows a structure beyond the excitation radiation distribution. This intensity structure is the more pronounced, the more clearly the emission power differs between the first and the second state.

By Irradiation of the excitation power above the threshold, the corresponds to the maximum of the fluorescence characteristic, less fluorescence stimulated than in the first state. In the microscope is the excitation radiation distribution with services over provided the threshold. Radiation of other wavelength, like it was previously required in the prior art is no more necessary. The microscope according to the invention or a device for execution of the method according to the invention therefore much easier to need in particular, only a single light source whose resolution is on the picture quality effect. Also, the chromatic requirements are regarding the coupling of the excitation radiation over the prior art clearly simplified, since no longer a relatively wide wavelength range must be covered.

The Microscope is tuned to the sample, since the irradiated excitation radiation distribution the Threshold for taken into account the line.

For the inventive approach is it about it In addition, it is not absolutely necessary that the sample areas, which with Excitation radiation power above the threshold illuminated are previously "normal" stimulated (as required for example in the STED or ESA approach is) when a sample is used after irradiation with over Threshold lying excitation radiation power permanently or at least for a certain amount of time a significantly reduced or even disappearing Excitability shows.

It is preferred that a A sample or a dye is used, the / at an excitation radiation power over the Threshold not luminescent, luminescence with others Properties radiates as the particular luminescence and / or modified, leading to reduced luminescence Absorption properties for excitation radiation having.

The resolution enhancement is thus according to the invention achieved that Sample areas, those with over Illuminates the threshold value of the excitation radiation intensity become, no or only little luminescence radiation for imaging contributes. Depending on the luminescence property of the sample in the second state, i.e. at and / or above the threshold lying excitation radiation power shows, this diminished contribution to the picture takes place in different ways. Shows the sample at all no more luminescence, is made from sample areas, in the second Condition are no longer detected even luminescence. Shows the sample in the second state, however, luminescence with changed optical properties, you will have a corresponding optical filtering, for example, with regard to spectral composition or polarization to hide. Show changed lives of the excited Lumineszenzzustände, a temporal filtering achieves a suppression.

Around to further narrow down the sample areas in the first state some of these sample areas select It is preferred to perform a confocal detection. By the reduction according to the invention of the luminescence-excited volume below the diffraction limit becomes a spatial resolution reached that surpasses that in normal confocal detection.

Essential is that not all illuminated with excitation radiation areas for luminescence contribute, but that with over the Threshold excitation radiation power illuminated sample areas in the picture are not included or only diminished, resulting automatically the resolution increase across from the irradiation of the excitation radiation results as a spatial reduction of the luminescent sample volume with respect to the excited sample volume is reached. It can be found in the picture optical structures in the Excitation radiation distribution were not present. The resolution of the Picture is over the resolution increased the excitation radiation injection addition.

In a particularly advantageous manner, a sample or dyes is used or the microscope according to the invention tunes it, which can be reset from the second state by irradiation of a reset radiation having other optical properties than the excitation radiation back to the (original) first state. The microscope has suitable means for this purpose. This means that the second state is an at least largely reversible state. By irradiation of excitation radiation power above the threshold value of the luminescent image hidden areas can then be excited after irradiation of the reset radiation back to normal luminescence in the first state. Such a further developed method or microscope designed in this way is then suitable for applications in which different regions of a sample are to be imaged several times. It is therefore preferred for such applications that a sample or a dye is used, the or in the second state, ie after illumination with Anreungsstrah If the radiation is above the threshold value, reduced sensitivity for excitation radiation is shown, wherein by irradiation of a reset radiation having other optical properties than the excitation radiation, the sensitivity reduction is at least partially reversed (and the first state is restored). At least the sample areas illuminated above the threshold excitation radiation power are irradiated with reset radiation. It is important in this case that the reset radiation, which transfers the sample back into the first state, does not contribute anything to effecting the increase in resolution, ie it can also be coupled into the sample in a very coarse or not at all structured manner.

In a particularly advantageous embodiment of the invention allowed in terms of their Lumineszenzeigenschaften back in the first Condition traceable sample imaging by scanning. The sample may be, for example with a spot, line or Multispot area are scanned, e.g. through a scanner device. Between two scan steps then each return radiation is irradiated, at the new scanning step each again excitation radiation under and over to be able to radiate the threshold value. As a result, an increased resolution is achieved in each scanning step, which is a clear improvement in resolution overall Picture delivers.

The Irradiation of the excitation radiation distribution thus generates a local Structure of fluorescence radiation. This is usually symmetrical with respect to Exciting radiation distribution. You want a certain fluorescent surface form, e.g. one point, thus advantageously parts of the fluorescent sample are hidden. This has no effect on the resolution of the desired surface shape possible, as you go for adequate distances between adjacent fluorescent regions.

For the suggestion the sample over or below the threshold value, in principle, two variants are possible. In a first variant is a principle diffraction limited Lighting, these being punctiform may be radially symmetric or a central maximum, as the known Airy function may have, but need not. In this In the form of the excitation distribution, it is preferred that the excitation radiation distribution diffraction-limited.

at The diffraction-limited excitation radiation distribution can be particularly advantageously a toroidal Use distribution in the core area of the toroidal distribution an excitation radiation power above the threshold, in marginal areas below the threshold. In the center circumscribed by the Torus, one then gets a punctate Luminescence distribution, which is significantly smaller than it is a diffraction-limited Airy slices would be. There, of course for the reasons of symmetry mentioned also on the outer edge the torus lying sample areas below the threshold lying Excitation radiation power are illuminated, there is luminescence radiation stimulated. To be a punctiform Fluorescence image, it is useful to hide these areas, for example, using a confocal detection, the only in the center circumscribed by the torus luminescence radiation lets happen. The luminescent center is still smaller than the (diffraction-limited) resolution the confocal detection per se allowed. The hiding can of course synonymous be done differently. For example, can the outdoor areas regarding the Lumineszenzanregbarkeit off targeted or the circumscribed Center can be switched on purposefully. The inventive method or microscope thus allows a resolution beyond the physical diffraction limit addition, while influencing the fluorescent area shape.

In a second variant is a flat, structured lighting used, the area also as one or more, diffraction-limited in one direction Line (s) can be formed. Subareas of the illuminated area have an excitation radiation power over and other subregions an excitation radiation power below the threshold value. A such areal Illumination allows a larger sample to raster very quickly. The same applies of course to a multi-spot arrangement, where multiple diffraction-limited lighting spots fall on the sample, which are so far apart from each other they are separated can be detected clearly separated from each other. When area structured sample illumination comes in particular a strip or Cross grid in question.

ever to sample Sample areas in the second state still have some residual luminescence demonstrate. This can either be due to the sample itself, or by migrating luminescent material between them differently lit sample areas, for example by Diffusion. The sample itself may e.g. annoying autofluorescence or contain other dyes that are not in a second, reversible Condition are transferable. As a result, one has a residual luminescence radiation in areas those with over Illuminates the threshold value of the excitation radiation power were. In principle, this residual luminescence can be replaced by a suitable Threshold in the detection can be suppressed.

The Using a on and for known lock-in technology suppresses this background radiation with improved picture signal. This is done either in sample areas that are in be placed in the second state or in sample areas, the remain in the first state, the excitation radiation according to a Intensity modulation of reference frequency and this reference frequency in the detection of the luminescence radiation using the lock-in technique. The microscope points to it a modulator and a downstream of the luminescence detector Lock-in amplifier on. This then separates the modulated signals and suppresses the unmodulated signals, which is a high-resolution detection even at the allows said residual luminescence, without that Peak level of the wanted signal decreases. As a result, the modulated wear Signals full to the high resolution Picture at.

A alternative possibility for suppression A residual luminescence is a difference method in which two images be subtracted from each other. A first picture is taken with excitation radiation power below the threshold, a second image with excitation radiation power recorded above the threshold. The difference between the separation of the two images causes the residual luminescence. Other possible approaches for suppression The residual luminescence use further properties of in variants The invention used samples or dyes, for example, the Evaluation of different luminescence lifetimes in the first or second state, the use of different optical properties the luminescence radiation in the first or second state o.ä.

Further, in advantageous embodiments, sample parts that lie outside the focal plane to be detected are hidden in order to increase the depth of field. This simultaneously decouples the axial from the lateral resolution. For this purpose, the sample is first transferred unstructured to the second state and then reset only in the focal plane. The same achieves a corresponding deep structuring of the excitation radiation intensity, with a minimum in the focal plane. For this one can use an arrangement according to the principles of DE 102 57 423 A1 modify suitably.

The initially mentioned inventive microscope realized in advantageous embodiments, one or more the above developments. Since the excitation radiation both to stimulate the luminescence as well as to prevent or reduce is used by luminescence, it is advantageously possible to use a provide only excitation radiation source, which the excitation radiation and a device is arranged downstream of which the beam profile imprinting a minimum, preferably of variable depth, wherein at least the power is below the threshold.

The The invention will be described below with reference to the drawings for example, even closer explained. In the drawings shows:

1a and 1b location-dependent power distributions for prior art methods,

2a - 2c Schematic drawings for various prior art dissolution enhancement methods;

3 a resolution-enhanced fluorescence microscope according to the prior art,

4 a fluorescence characteristic of a sample,

5 a schematic drawing of the resolution increase according to the invention,

6a / b one- or two-dimensional power distributions, such as occur in the increase in resolution according to a first and second method,

7 a representation similar to the 6a / b for a third method,

8a / b representations similar to the 6a / b for a fourth and fifth method,

9a / b a schematic drawing for a first resolution-enhanced microscope,

10 a power distribution similar to that 6a to illustrate the operation of the microscope of 9 .

11a / b a characteristic similar to the 4 and a power distribution to illustrate a second resolution enhanced microscope,

12 a schematic representation of the second microscope,

13 a schematic representation of a third resolution enhanced microscope,

14 a schematic representation of the operation of the third microscope,

15a / b Schematic representations of parts of a fourth microscope,

16a a distribution of the excitation or reset radiation in a pupil of the fourth microscope,

16b a multispot distribution realized with the fourth microscope,

17 a fluorescence distribution in the fourth microscope and

18 a distribution of the excitation or reset radiation in a pupil of a variant of the fourth microscope.

following be in procedural explanation as well as in microscope descriptions different embodiments to the resolution-enhanced Luminescence microscopy shown. This is purely exemplary using various fluorescence microscopes or fluorescence microscopy methods. Naturally are the embodiments described below, d. H. process forms and microscopes, too, for different luminescent substances used, for example for phosphors Samples or dyes. As far as subsequently spoken by a dye this is only to be understood as an example. Instead of of a dye used for preparation a sample can be used, of course, a directly fluorescent (or phosphorescent) substance as a sample, whereby a Dye addition is dispensable. Also, additional dyes can be used that show other qualities and not in a second one Condition can be brought. Next you can for individual Process forms or microscopes described features for others described method forms or microscopes are used, so that too here unspecified combinations are possible.

4 shows by way of example a fluorescence characteristic 17 for a dye used in the invention. This fluorescence characteristic 17 represents the power L of the emitted fluorescence radiation F as a function of the power L of the excitation radiation A again. As can be seen, the characteristic increases 17 largely linear, at least monotone, up to a maximum 18 and then falls above a threshold 19 for the excitation radiation power again. Recognizable are two state regions, each to the left and right of the threshold 19 or the maximum 18 lie. Will an excitation radiation power below the threshold 19 irradiated, the dye is primarily in a first state 5 , In this he can emit fluorescence radiation. Will the excitation radiation power above the threshold 19 increases, a rearrangement of the dye molecules in a second state 6 instead of. Dye molecules in the second state 6 either can not fluoresce, emit fluorescence radiation with altered optical properties, and / or have one opposite the first state 5 changed absorption property. The altered optical properties for fluorescence emission or absorption may relate to the spectral composition, the polarization and / or the lifetime of the fluorescence radiation. As a result, in the second state region, the power of fluorescence radiation F falls, considering the type of fluorescence radiation in the first state 5 will resign, again.

The invention now sets the characteristic 17 to the effect that the irradiation of the excitation radiation A only a part of the sample in the first state region (or short first state) 5 but place another part of the sample in the second state region (or second state for short) 6 brings. Areas with excitation radiation power above the threshold 19 were / are illuminated, emit thus the specific fluorescence radiation only reduced. It is advantageous if the second state 6 is a reversible state. This means that at least after a certain time or after active action once with excitation radiation power above the threshold value 19 irradiated dye again the properties of the first state 5 shows. An example of a dye for which the inventors have a characteristic according to the 4 The substance referred to as Dronpa, described in the publication R. Ando et al., Regulated almost nucleocytoplasmic shuttling observed by reversible protein highlighting, Science, November 19, 2004, Vol. 306, pp. 1370-1373 The dye is available from Amalgaam, Woburn, Massachusetts, under the name Dronpa Green, Code No. AM-V0071, and has heretofore been used in connection with the fluorescence labeling of molecules whose transport properties are to be examined by cell membranes. The dye Dronpa can be excited at a wavelength of 488 nm, and after excitation in the second state 6 by radiation at a wavelength of 405 nm back to the first state 5 to be led back.

In combination with a suitable excitation radiation distribution when illuminating the sample, the use of the dye with the characteristic of the 4 or a corresponding characteristic to a significant increase in the resolution. In this case, in contrast to the prior art, only radiation having a wavelength is irradiated, and only this radiation contributes to the resolution. The switchback may be those embodiments in which the dye is not spontaneously in the first state 5 returns, be achieved by irradiation of a reset radiation R, but the Auflö increase is irrelevant.

With the excitation radiation A is in addition to the excitation of the dye molecules from the base level out in the first excited level (first state 5 ) also the prevention of fluorescence or depopulation (second state 6 ) accomplished. During depopulation, the dye molecules are rearranged from the first state to the second state. The two states have different optical fluorescence properties from each other. After the excitation with the excitation radiation A, the high-resolution fluorescence image can be obtained immediately, wherein high-resolution refers to the optical resolution of the excitation radiation A. If the dye from the second state does not spontaneously return to the first state, a reset, for example by means of a reset radiation R, which has different optical properties from the excitation radiation A, is actively accomplished. The corresponding scheme shows 5 ,

Originally, for example, the sample P to be examined is in the first state 5 in which the power of the fluorescence radiation F increases with the power of the excitation radiation A. The irradiation of the excitation radiation A is now carried out in such a way that in individual sample areas B2 the excitation radiation power is above the threshold value 19 lies. In other areas B1 it is below. Thus, some sample areas are in the first state 5 , others in the second state 6 , depending on whether the power of the excitation radiation above or below the threshold value 19 lies. 5 schematically shows the transfer of the sample by irradiation of excitation radiation A in sample areas B1 in the first state 5 (right part of the lower half of the 5 ) as well as sample areas B2 in the second state 6 (left part). Sample areas B2 in the second state 6 can no longer efficiently fluoresce, especially if an excitation radiation power is well above the threshold 19 has been used. The term "fluorescence" refers to fluorescence radiation with certain properties, so it is quite possible that even in the second state 6 Fluorescence radiation is emitted, but with different properties than in the first state 5 , Either the absorption properties and / or the fluorescence properties of the sample change in the second state.

Areas B1 of the sample in the first state 5 remain, emit after irradiation of excitation radiation A continues the same fluorescence radiation F. irradiation of reset radiation R (in 5 this process is illustrated by a corresponding arrow) brings the entire sample P back to the first state 5 , reverses the division into two different fluorescent sample areas B1 and B2.

The division of the sample into sample areas B1, which are in the first state 5 and sample areas B2, which are in the second state 6 Now, it is possible to limit the emission of fluorescence radiation F to a volume of the sample P which is smaller than the original volume illuminated with excitation radiation A. Thus, it is possible to achieve a structure in the fluorescent image that was not originally present in the excitation radiation distribution. The reason for this is the cleverly used nonlinearity of the characteristic curve 17 ,

The 6a and 6b show two procedural forms, how the excitation radiation distribution can look to increase the resolution. Both figures show on the left a one-dimensional section through the excitation radiation distribution or fluorescence radiation distribution. On the right in each case the two-dimensional image of the fluorescence radiation distribution can be seen in plan view.

In 6a a punk-shaped distributed excitation radiation A is irradiated. The distribution here is a diffraction-limited Airy distribution. As a result, fluorescence radiation F is excited in the dye with an ellipsoidal donut distribution, which is represented in the two-dimensional image as a laterally fluorescent, circular ring. Only the sample areas B1 fluoresce. Axially is a fluorescent elliptical ring. This sample area B1 is smaller than the diffraction limit of the excitation radiation allowed.

If you shine on the other hand, as in 6b shown, a likewise diffraction-limited donut-shaped distribution of the excitation radiation A, results in a radiating sample area B1, which corresponds to a reduced Airy distribution, which is surrounded by an annular fluorescent region. In the sectional view, three peaks therefore appear in the distribution of the fluorescence radiation F as sample areas B1. Again, the point spread distribution 16 smaller than the diffraction limit allows. The improvement of the resolution over the normal point spread distribution depends largely on the combination of the characteristic 17 and power distribution of the excitation radiation A from.

In addition, for masking the annular outer area, it is possible to perform a confocal detection which selects the middle peak by appropriately setting a confocal detection volume D. This is schematically as a third method form in 7 shown. The now still detected fluorescent sample area B1 then consists of the middle peak. Overall is eg an imaging volume B (dotted line in the right-hand illustration of FIG 7 ), which is two and a half times narrower than the conventional spot-spread distribution, and even 1.7 times narrower than a confocal point-spread distribution. For a donut-shaped excitation distribution, as the solid curve in the right representation of 6b respectively. 7 shows, the resulting distribution of the imaging volume B and the mean peak of the point-spread distribution, respectively 6b radially symmetrical. If two illumination spots are used, a maximum increase in resolution is achieved along the line connecting the spot maxima.

Alternatively, the outer fluorescent ring in 6b by a targeted switching back of the dye from the second state 6 in the first state 5 hidden in the confocal detection volume D. The procedure / principle is then as follows: First, it is excited with excitation radiation so that the sample section to be examined completely reaches the second state. Subsequently, reset radiation R is irradiated with a distribution corresponding to the confocal detection volume D, and thus a part of the sample is switched back. Upon repeated excitation with the excitation radiation distribution A according to 7 Then only the lying in this sample part imaging area B in the first state and emits fluorescence radiation F in the form of the middle peak, as he also in the variant of 7 remains. The annular outer region remains in the second state and does not fluoresce.

In a fourth method form, a structured surface illumination, as they are, for example, in FIG 8a is shown. An excitation radiation A distributed sinusoidally in the x-direction leads to fluorescence radiation F in the valleys of the sinusoidal power distribution. The fluorescent sample areas B1 are strip-shaped. However, the strips are significantly narrower than the strips of the sinusoidal distribution of the excitation radiation A. It is advantageous here that the sinusoidal distribution of the excitation radiation A as minima has no zeros or must have. The resulting distribution of the fluorescence radiation F does indeed depend on the power of the excitation minima in terms of their height, but not in terms of their width. The depth of the minima of the sinusoidal excitation radiation distribution is thus uncritical for the width of the stripes and thus for the resolution. In the example case, the period of the sinusoidal strip distribution of the excitation radiation A is set at the cut-off frequency of the resolution. As a detector, for example, a matrix detector is used. The stripe width and thus the resolution is improved by a factor of 6 compared to the cutoff frequency of the resolution. A high-resolution image is now obtained by shifting the stripe pattern.

alternative can instead of a strip-shaped wide-field illumination also a burning lighting be used, with the line along its longitudinal axis (e.g., sinusoidal) is modulated so that the Performance in line sections above the Threshold and in line sections below the threshold. The scanning movement is perpendicular to the line and along the line. The detector is a suitable high-resolution line detector.

8b shows a development of the fourth method of the 8a , In this fifth method, the distribution of the resolution radiation A is structured in both lateral axes, ie in the sample plane, so that a matrix-shaped illumination spot distribution is given. This is in the sectional view of the left half of the 8b represented one-dimensionally. Perpendicular to the optical axis, a matrix of bright excitation spots is shown on the sample. Each illumination spot is now assigned a detector. A shift of the spot pattern relative to the sample also allows the spaces between the points to be measured. The area in which the displacement of the matrix-shaped pattern is to take place is in 8b represented as area G1. By scanning the spot matrix, for example as is known with a conventional laser scanning microscope, overall the entire area G2 is composed of individual areas G1. This parallelization achieves a very high scanning speed. At the same time, as the narrow peaks of the fluorescence radiation F show in the sectional view, the resolution is increased beyond the resolution of the excitation radiation distribution.

Even though so far only the lateral resolution mentioned was, is also the axial resolution for the embodiments improved.

9a shows a laser scanning microscope, which can realize each of the explained method forms. For the procedural forms 1 to 3 is the laser scanning microscope 20 z. B. formed as a single-point scanning microscope. It has an excitation module 21 , a microscope module 22 and a detector module 23 on. In the microscope module 22 there is a sample 24 in the focus of a lens 25 , in the direction of illumination a tube lens 26 upstream. In front of this optics is a scanning optics 27 that together with a scanner 28 a scan of the sample 24 by moving the focus point on the sample allows. A main color divider 29 couples the radiation from the excitation radiation module 21 in the microscope module 22 and disconnects from the microscope module 22 from the sample 24 absorbed radiation to the detector module 23 from.

The excitation module 21 has a light source 30 whose radiation is above the main color splitter 29 to the focal point in the sample 24 is bundled. The focus point of the sample 24 excited fluorescence radiation F is from the objective 25 collected and at the main paint divider 29 due to the changed compared to the excitation radiation A spectral properties to a Pinholeoptik 31 decoupled, the one pinhole 32 and an (optional) block filter 33 are downstream. A point detector 34 detects the power of the fluorescence radiation F at the focal point. The detection can additionally be spectrally resolved, polarization-resolved and / or time-resolved. The signals of the detector 34 be from a controller 35 read out the total operation of the laser scanning microscope 20 controls.

To increase the resolution, the irradiation of the excitation radiation A takes place in the focal point 24 with a certain power distribution. To specify and adjust this is a suitable structure in the excitation module 21 intended. The construction of the 9a has a beam splitter 36 on, the 50% of the beam power of the light source 30 couples out. In the embodiment, the beam splitter is 36 around a polarization beam splitter. The sub-beams split up in this way are subsequently transferred via a further beam splitter 40 superimposed again into a common beam, wherein the intensities and phases of the partial beams are adjusted beforehand. In addition, in a partial beam is a phase element 39 placed, which can be formed either as a solid phase element or as a variably adjustable phase element. The phase element 39 has z. B. a phase-changing region 44 as well as a phase neutral range 49 so that a donut-shaped power distribution is generated in the beam. After superposition of the two partial beams after the beam splitter 40 one obtains excitation radiation A according to 10 , The structure of the phase element 39 is exemplary in 9b shown. To adjust the power components of the two partial beams before the superposition, variable attenuators are still present in each partial beam 37 respectively. 38 posed.

The result is a distribution of the power of the excitation radiation A, as in 10 is shown as a sectional view along an axis perpendicular to the optical axis extending x-axis. The distribution has a minimum 45 on, its depth by adjusting the attenuator 38 respectively. 37 , that is variably adjustable by changing the relative power of the two partial beams, z. B. by the controller 35 , The adaptation of the minimum 45 allows the intensity of the also in 10 set fluorescence radiation F optimal.

The minimum adjustable in depth 45 can also be generated by the use of a variably adjustable phase element without formation of partial beams. For such a phase element is a matrix of liquid crystals in question, in which the phase of each pixel is adjustable. Of course, other means for generating a donatförmigen beam distribution with adjustable minimum 45 possible.

For an optional downshift of the sample areas B2 from the second state 6 in the first state 5 may be a reset radiation source 41 be provided, which has a third divider 42 in the excitation beam path of the excitation module 21 is involved. However, this design is optional, which is why the presentation of the 9a also shows a dashed representation of these elements. The reset radiation source 41 is only required if the dye of the sample 24 not spontaneously returns to the first state, but an active reset by optical reset radiation R requires. However, reset radiation can also be used elsewhere 24 be applied, for example, by an irradiation oblique to the optical axis of the lens 25 with the aid of a reset beam source mounted laterally by the laser scanning microscope.

The microscope of 9a can perform the initially described method forms. The third method requires the use of the pinhole 32 and Pinholeoptik 31 in the detector module 23 , Whenever confocal detection is not required, the necessary components ( 31 . 32 ) accounted for.

11a shows a further possibility for improving the procedure according to the invention. The transition of the sample in the partial area B2 from the first state to the second state 6 can take place incomplete. This is the case, for example, when the fluorescence characteristic 17 after the maximum value 18 , ie for powers of the excitation radiation A above the threshold value 19 not drop steeply enough. In the example of 11a goes the characteristic 17 even at high excitation radiation powers not back to zero. This leads to a residual fluorescence 47 in the sample areas B2, in the second state 6 are, as the reduced share 46 not the maximum value 18 equivalent. If one then the fluorescence radiation F then with the microscope according to 9a For example, in the realization of the third method form, the imaging volume B would be composed of a low-resolution base S and a high-resolution distribution V. This is in 11b schematically shown hatching the base S and shows the high-resolution distribution V black.

The same effect can occur if there is a strong diffusion of the fluorescent material, for example the dye molecules, in the sample. Even then, a residual fluorescence results 47 in the sample areas B2, but not due to a comparatively weak slope 17 but by diffusion of the fluorescent material.

The base S can be suppressed by suitable image recording. A second embodiment of a laser scanning microscope realizing this 20 is exemplary in 12 shown. With the construction according to 9a Matching components are provided with the same reference numerals, so that it can be dispensed with their repeated explanation.

In contrast to the construction of the 9a is now the detector 34 with a lock-in amplifier 48 connected, this connection in 12 merely an example of the control unit 35 is realized. Of course, a direct connection is possible. Also, the lock-in amplifier 48 in the control unit 35 to get integrated. The lock-in amplifier 48 is with a frequency generator 49 connected, the minimum 45 modulated in its amplitude. For this purpose, the power of one of the two partial beams in the excitation module 21 suitably modulated what z. B. by controlling one of the two attenuators 37 or 38 happens. In the presentation of the 12 is the control line for the attenuator 38 dashed to indicate this option. Of course, the control of the attenuator by the control unit 35 take place, which is the signal of the frequency generator 49 evaluated accordingly. As mentioned earlier, the lock-in amplifier can 48 and / or the frequency generator 49 Components of the control unit 35 be.

The from the frequency generator 49 The frequency provided for modulating the minimum will also be the lock-in amplifier 48 fed. By the modulation of the minimum 45 the signal changes from the sample areas in the first state 5 while signals from pedestal S remain constant. The lock-in amplifier 48 separates out the modulated signals and rejects signals from the pedestal S. As a result, the distribution V is separated from the pedestal S, and the high-resolution detection is also at an incomplete transition to the second state 6 reached.

recycled one directs the direct component of the signal, one reaches an additional one Image information that is not high resolution. This image information can e.g. on different fluorescent sample areas / properties Respectively. The high and low resolution may be superimposed arbitrarily encoded (e.g., color coded) being represented.

As an alternative or in addition to this lock-in detection, two images can also be recorded. A first picture is taken with excitation radiation distribution with minimum 45 , the second without minimum 45 recorded. In order to separate the high-resolution distribution V from the base S, only the images are then subtracted from each other (first image minus second image). The result image only contains the information with the high-resolution distribution V. Further possibilities for suppressing a low-resolution socket can be the use of the different optical properties of the first state 5 and the second state 6 be, for example, if the lifetime of the two states differ. Also, one can use a different characteristic of the emission spectrum, or a different characteristic of the absorption spectrum between the first state 5 and second state 6 ,

13 shows a further modification of the laser scanning microscope 20 of the 9a for the realization of one of the method forms with linear illumination. Again, with the 9a matching elements with the same reference numerals, so that in turn their description can be dispensed with. Next only the center rays are drawn.

The now coherent light source 30 radiates its radiation through a grid 50 , which produces two 1st diffraction orders, o + and o-, as well as a 0th diffraction order. The grid is a cylindrical lens 50a downstream, which is an example of an anamorphic element here. A line perpendicular to the optical axis is focused on the sample.

The diffraction orders fall when focusing grazing to each other and thus arrive in the sample to interference. The Talbot effect creates a so-called Talbot grid along the optical axis OA. This is schematically in the upper half of the 14 shown. The +1, the 0 and the -1. Order fall in the sample 24 at different angles and the interferences produce the Talbot structure known to those skilled in the art.

The effect within the sample is shown in the lower half 14 which shows a depth section (x, z-plane) through the sample. At high-intensity points (black areas B2), the sample enters the second state 6 That is, a depopulation of the excited states is achieved by switching the dye molecules. In areas B1 of low intensity, the sample remains in the first state 5 (hatched and white areas).

The detection takes place with a line detector 34 who is in the construction of the 13 exemplary than with the control unit 35 is shown united. The line detector 34 is again (optional) a block filter 33 presented. Next is the line decode 34 a (in 13 not shown, because directly on the line detector 34 applied) disposed along the x-axis slit diaphragm to cause the required confocal selection of a depth plane and thereby detect only the hatched areas.

The scanner 28 allows movement of the excitation radiation distribution across the sample 24 perpendicular to the optical axis, ie along the x and y axes. Optionally, with the reset radiation source 41 for example, a homogeneous line along the x-axis in the sample 24 be generated, which is a switching of the dyes from the second state 6 in the first state 5 allows, like this 5 illustrated. This line may also have a lighting pattern matching that of the sample areas B2 in the second state 6 is because only these areas need to be reset.

The laser scanning microscope according to 13 So realized the method form according to 8a , The depth of the minima 45 can be done by the quality of the grid or the individual attenuation of the individual orders.

15a shows an arrangement for implementing the method according to 8b , being by laser scanning microscope 20 here only the excitation module 21 shown is because the microscope module 22 otherwise has the structure already described. From the light source 30 be by means of a first beam splitter 53 generates two partial beams, spaced here perpendicular to the plane of the drawing. A second beam splitter 52 generates a total of four parallel partial beams. 15b shows the arrangement in a rotated by 90 ° degree section. Altogether there are four partial beams 54 . 55 same intensity (the center rays are drawn). In the drawing, two partial beams each take the same path, so that they are hidden.

The partial beams are transmitted through lenses into a pupil of the microscope module 22 focused so that within a boundary 56 the pupil the in 16a shown point distribution results. The four partial beams 55 . 54 take in the corners of a square. In the sample 24 the sub-beams reach the interference, whereby perpendicular to the optical axis a multi-spot pattern 58 according to 16b on the test 24 arises. The depth of the zeros of this multi-spot pattern 58 For example, by changing the intensities of the respective diagonally opposite rays 54 . 55 be varied. Suitable adjustable attenuators (not shown) for the partial beams are provided for this purpose.

In the high intensity areas (bright areas B2) in the multi-spot pattern 58 of the 16b passes the sample 24 in the second state 6 , Fluorescence radiation F arises in the areas B1 of low intensity (dark areas). The fluorescence radiation F is then as usual at the main color splitter 29 separated. The detector 54 is a matched to the spot pattern matrix detector, optionally with upstream Pinholemaske. The detector elements are aligned with the areas B1. The scanner 28 in turn allows the displacement of the excitation radiation distribution across the sample 24 in the x and / or y direction. 17 shows in the picture below the multi-spot pattern 58 and above that the fluorescence pattern excited thereby 59 ,

From the reset light source 41 can another beam 57 be focused in the pupil, making it a homogeneous surface in the sample 24 illuminates. This results in a switching back of the fluorescence molecules from the second state 6 in the first state 5 , as already explained between two scan steps and possibly also within a Schanschrittes.

The reset beam 47 but can also have a lighting pattern that corresponds to the areas B2, since only these must be reset. For this purpose, the beam 47 analogous to the partial beams 54 . 55 also divided into four sub-rays and imaged in the pupil, so that within the boundary 56 the pupil, for example, the in 18 shown distribution results. In addition, the optical axis of the partial beams 57 the reset radiation R relative to the optical axis of the excitation radiation A tilted such that the resulting in the sample minima of the multi-spot pattern 58 the excitation radiation A coincide with the maxima of the multi-shot reset pattern. The embodiment according to FIG 15a in the variant of 18 is an exemplary example that by appropriate structuring of the reset radiation, a sample or dye-sparing reset takes place only in the areas B2.

It is important that the sample is divided by appropriate irradiation of excitation radiation A in areas B1 in the first state and in areas B2 in the second state. The transfer of sample areas to the second state can therefore also be achieved by selecting a depth plane of the sample (cf. 14 ), so that sample areas B1 can not contribute to crosstalk of adjacent detector elements / regions.

For the method or microscope described above in different variants, the adjustment of the power of the excitation radiation A is dependent on the fluorescence characteristic of the sample with special consideration of the local distribution of the excitation radiation A of importance. In particular, the depth of one or more minima determines 45 and the height of the corresponding maxima of the excitation radiation of the distribution the achievable resolution increase. An optimization of the adjustment of the local power distribution of the excitation radiation A can be realized in two ways: With knowledge of the characteristic curve 17 , for example, the control unit 35 is provided according to, is calculated for a given local excitation distribution of the optimum power level and adjusted accordingly, for example by outputting a setpoint or even direct control of the excitation module 21 ,

Alternatively or additionally, the actual resolution can be determined on a test preparation or a reference point of the sample to be examined, and the optimum power level for a given excitation distribution can be determined in an interactive process and then used to image the sample 24 be used.

Instead of a pure power scaling, of course, the distribution can be changed, for example, by appropriate adjustment of the phase element 39 ,

The fluorescence characteristic of the sample 24 For example, the fluorescence characteristic of a dye can be determined by examination on a test preparation or a reference site of the sample already under later image recovery conditions. For this purpose, the power of the detected fluorescence radiation F is determined as a function of the power of the excitation radiation A at a point or an area of the sample. For a particularly good setting, it is favorable if the fluorescence characteristic, ie the characteristic 17 does not depend on the concentration of the fluorescent material, for example the dye (in contrast to the emitted power of the fluorescence radiation F). The control unit 35 of the scanning microscope 20 can do this from the determined curve 17 or a supplied or stored in a memory curve 17 Automatically determine the optimal performance and adjust accordingly.

The iterative determination is always advantageous if the fluorescence properties are strongly sample-dependent and a reference point on the sample for determining the fluorescence characteristic is present or can not be found. In this case, one can either optimize the performance at a limited area inside or outside the sample field of interest or by means of imaging in the entire sample field to be imaged or by the control unit 35 let execute. The excitation power is optimized on the basis of a quality criterion of the resolution (for example the contrast in a selected location or the extension of the frequency range transmitted in the image in the Fourier space). It is generally sufficient to adjust the power in the minimum of the excitation distribution or a spatially broad distributed underground performance.

The Excitation radiation A and the reset radiation R can be generated from a single radiation source. The radiation source can either directly switchable or with a downstream selection arrangement be provided.

Claims (30)

Dissolution enhanced luminescence microscopy method in which - a sample (P, 24 ) is excited by the irradiation of excitation radiation (A) for the emission of specific luminescence radiation (F) and an image of the luminescent sample (P, 24 ), wherein - the luminescent sample (P, 24 ) from a first luminescence state ( 5 ), in which the excitability for the emission of the determined luminescence radiation (F) with increasing excitation radiation power up to a maximum value ( 18 ), which corresponds to an excitation radiation power threshold ( 19 ), rises to a second luminescence state ( 6 ), in which the sample (P, 24 ) compared to the first state ( 5 ) has reduced excitability for emission of the determined luminescence radiation (F), the sample (P, 24 ) by irradiation of excitation radiation power above the threshold value ( 19 ) in the second state ( 6 ), - the sample (P, 24 ) in sub-areas (B1) in the first state ( 5 ) and in adjacent sub-areas (B2) in the second state ( 6 ) is offset by the irradiation of excitation radiation (A) with an excitation radiation distribution which at least a local power maximum above the threshold ( 19 ) and at least one local local minimum ( 45 ) below the threshold value, - the image of the luminescent sample (P, 24 ) Sample areas (B1) in the first state ( 5 ) and sample areas (B2) in the second state ( 6 ), wherein the image of the luminescent sample (P, 24 ) predominantly sample areas (B1) in the first state ( 5 ) and thereby the image has a spatial resolution which is increased in relation to the excitation radiation distribution. Method according to Claim 1, in which a sample (P, 24 ), which after transfer to the second state ( 6 ) back to the first state ( 5 ), either through active intervention kung or spontaneously. Method according to Claim 2, in which a reset radiation (R) is irradiated, which irradiates the sample (P, 24 ) from the second state ( 6 ) in the first state ( 5 ) and which has different optical properties from the excitation radiation (A). Method according to one of the preceding claims, in which the image is obtained by scanning the sample (P, 24 ) with a spot, line or multi-spot pattern ( 58 ) is recovered and in particular between two scan steps reset radiation (R) is set. Method according to one of the preceding claims, in which a sample (P, 24 ) which is at an excitation radiation power above the threshold ( 19 ) does not luminesce, emits luminescence radiation having properties other than the particular luminescence radiation (F) and / or has altered properties leading to reduced luminescence. Method according to one of claims 1 to 5, wherein the excitation radiation distribution diffraction-limited. The method of claim 6, wherein a torus distributed Excitation radiation (A) is used, and preferably in addition the detection of the luminescence radiation (F) outside of the torus Sample areas are hidden. Method according to one of claims 1 to 5, wherein the excitation radiation distribution (A) a linear or areal structured sample illumination, in particular according to a Strip or cross lattice, and in which a spatially resolved detection used for image acquisition. Method according to one of the preceding claims, in which a confocal detection (D) of in the first state ( 5 ) carried out sample areas (B1). Method according to one of claims 1 to 9, in which a depth resolution increase along the optical axis (OA) takes place by the power minimum ( 45 ) below the threshold ( 19 ) and the power maximum above the threshold ( 19 ) are adjacent along the optical axis (OA). Method according to one of Claims 1 to 9, in which a depth resolution increase takes place along the optical axis, by first excitation radiation (A) having a power above the threshold being irradiated in a depth region containing the focal plane, re-radiation (R) being radiated into the focal plane, and Subsequently, the excitation radiation (A) is irradiated with the excitation radiation distribution, wherein the power minimum ( 45 ) is in the focus area. Method according to one of the preceding claims, in which either in sample areas (B1), the first state ( 5 ), or in sample areas (B2), the second state ( 6 ), the excitation radiation power is modulated according to a reference frequency, and this reference frequency is used in the detection of the luminescence radiation by way of the lock-in technique. The method of claim 12, wherein additional from a DC signal component (S), a low-resolution image is won. Method according to one of the preceding claims, in which a fluorescent sample (P, 24 ), in particular a sample provided with at least one fluorophore (P, 24 ). Method according to one of the preceding claims, in which additionally an image with excitation radiation distribution without power minimum ( 45 ) is taken below the threshold and used for subtraction. Method according to one of the preceding claims, in which a luminescence characteristic ( 17 ), which indicates the emitted luminescence power as a function of the excitation radiation power, evaluates the threshold value ( 19 ) and / or the excitation radiation distribution is selected as a function of the characteristic. Method according to one of the above claims, wherein the excitation radiation distribution iteratively under evaluation the image resolution is optimized. Microscope for enhanced resolution luminescence microscopy, comprising: - means ( 21 . 22 ) for excitation of luminescence, the excitation radiation (A) on the sample (P, 24 ) and thus the emission of certain luminescence radiation (F) in the sample (P, 24 ), - means ( 22 . 23 ) for obtaining an image of the luminescent sample (P, 24 ), Wherein the means for excitation irradiate the excitation radiation (A) with a specific excitation radiation distribution which has at least one local power maximum which is above a threshold value ( 19 ), and at least one local local minimum ( 45 ), which is below the threshold ( 19 ), the threshold value ( 19 ) two luminescence regions ( 5 . 6 ) of the sample (P, 24 ), a first, at excitation radiation powers below the threshold ( 19 ) present state region ( 5 ), in which the excitability for the emission of be tuned luminescent radiation (F) with increasing excitation radiation power up to a maximum value ( 18 ), which at the threshold ( 19 ) is reached, and a second at and / or after excitation radiation powers above the threshold ( 19 ) present state region ( 6 ), in which the sample (P, 24 ) one opposite the first region ( 5 ) has reduced excitability for emission of the determined luminescence radiation (F), and - wherein the image acquisition means comprises sample regions (B1) in the first region ( 5 ) with excitation radiation power below the threshold value ( 19 ) and sample areas (B2) in the second region ( 6 ) with excitation radiation power above the threshold ( 19 ), wherein the image of the sample is predominantly sample regions (B1) in the first region (FIG. 5 ) and thereby the image has a spatial resolution which is increased in relation to the excitation radiation distribution. Microscope according to Claim 18, having an excitation radiation source ( 30 ), which emits the excitation radiation (A), and one of the excitation radiation source downstream device ( 36 - 40 ) which emits an excitation beam having a beam profile with the power minimum ( 45 ), one below the threshold ( 19 ) has lying excitation radiation power. Microscope according to Claim 18 or 19, characterized in that a scanner device ( 28 ) the sample (P, 24 ) with excitation radiation (A) scans. Microscope according to claim 20, wherein the excitation radiation distribution has at least one diffraction-limited point image. A microscope according to claim 21, wherein the excitation radiation distribution illuminates at least one line-shaped or strip-like or grid-shaped sample area with excitation radiation (A), wherein the excitation radiation power in parts of the grid or grid exceeds the threshold value (A). 19 ) and in parts of the strip or grid below the threshold ( 19 ) lies. Microscope according to one of claims 18 to 22, wherein the excitation radiation distribution the local power minimum ( 45 ) and the local power minimum along the optical axis (OA) are adjacent. Microscope according to one of claims 18 to 23, characterized by means ( 41 . 42 ) for resetting the sample ( 24 ) from the second ( 6 ) into the first state region ( 5 ), which is a beam source ( 41 ), which emits reset radiation (R) with optical properties different from the excitation radiation (A), and an optical system ( 42 ) for coupling the reset radiation (R). Microscope according to one of Claims 18 to 24, characterized by an intensity modulator ( 37 . 38 . 49 . 35 ) which modulates the distribution of the excitation radiation (A) with a reference frequency, and a lock-in amplifier provided in or connected to the image acquisition means ( 48 ), which evaluates the reference frequency and reduces background radiation (S). Microscope according to claim 25, having a DC signal output to gain an additional low resolution Image. Microscope according to one of Claims 18 to 26, having a control device ( 35 ), which drives the microscope for carrying out one of the methods according to one of claims 1 to 17. Microscope according to one of Claims 18 to 27, having a control device ( 35 ) which activates or reads out the means for excitation and the means for image acquisition and adjusts the excitation radiation distribution to optimize the image resolution. Microscope according to Claim 28, in which the control device ( 35 ) has a memory or data input device via which the control device has a luminescence characteristic ( 17 ), the luminescence characteristic ( 17 ) reflects the emitted luminescence power as a function of the excitation radiation power and wherein the control unit ( 35 ) from the luminescence characteristic ( 17 ) the threshold ( 19 ) and / or adjusts the excitation radiation distribution as a function of the characteristic. Microscope according to Claim 28, in which the control device ( 35 ) from the image of the sample (P, 24 ) determines the resolution and adjusts the excitation radiation distribution iteratively to optimize the image resolution.