Script Bio 407 Applied Microscopy Light microscopy

Transcription

1 Center for Microscopy and Image Analysis Dr. José María Mateos Center for Microscopy and Image Analysis Winterthurerstrasse 190 CH-8057 Zurich Phone Script Bio 407 Applied Microscopy Light microscopy "How to look at one tree in the forest, and why?" Left: drawing of dendritic spines of pyramidal cells by S. Ramón y Cajal (1896). Right: image composition of theoretical and practical microscopy elements that will be discussed during the course. BIO 407 April 16 th - May 8 th 2013 Aim Bio 407 course aims to impart knowledge about light and electron microscopy in theory and practice. Theoretical basics will be embedded in full-time and hands-on sessions. We will use as a model system the synaptic contacts of the central nervous system.

2 "How to look at one tree in the forest, and why?" Understanding the complexity of the synaptic circuit is one most challenging tasks of current biology. Neurons communicate with each other by tiny structures called synapses. These contacts are established during development but, can appear and disappear under processes as learning and memory. Thus, synapses are considered plastic structures which remodel the synaptic circuit and allow us to learn and remember events. However, synapses are close to the resolution limit of light microscopy and the visualization of these structures is a demanding task. During this course we will image neurons and synapses by fluorescence and electron microscopy and discuss which microscopic approach is more appropriate to use according to which scientific question. Cover image composition: Diffraction patterns; GFP-positive neuron in the hippocampus (Mouse line courtesy Prof. P. Caroni, FMI, Basel; Airy disks and resolution; Laser lines; Filter cube; Hippocampal neurons imaged in a multiphoton microscope; CCD detector system; Dendritic spines; Phase contrast rings; cerebellum (sample courtesy Prof. J. Loffing, University Zurich); Emission spectra; Cells stained with different fluorescent markers. This script has been written as a theoretical summary of Bio 407- light microscopy part. The information in the text and many figures have been obtained from Fundamentals of light microscopy and electronic imaging D.B. Murphy, Wiley-Liss; Light and electron microscopy E. M. Slayter and H.S. Slayter, Cambridge University Press, from scripts and lectures from Dr. Urs Ziegler and Dr. José María Mateos, Center for Microscopy and Image Analysis, University Zurich and also other sources. 2

3 INDEX 1. INTRODUCTION 4 MAGNIFICATION VERSUS RESOLUTION 4 2. FUNDAMENTALS OF LIGHT MICROSCOPY 5 BASIS OF A MICROSCOPE- TYPES 5 APERTURE, IMAGE PLANES AND KÖHLER ILLUMINATION 6 OBJECTIVES AND IMAGING ERRORS 7 LAMPS AND SPECIMEN ILLUMINATION 9 3. IMAGE FORMATION BY LENSES AND LIGHT 11 HOW IS AN IMAGE FORMED IN A MICROSCOPE 11 LIGHT AS A PROBE OF MATTER 11 DIFFRACTION AND IMAGE FORMATION 12 RESOLUTION PHASE CONTRAST AND DARK FIELD MICROSCOPY 16 PHASE CONTRAST MICROSCOPY 17 DARK FIELD MICROSCOPY DIFFERENTIAL INTERFERENCE MICROSCOPY 18 DIC COMPONENTS 19 SETUP OF A DIC MICROSCOPE AND FORMATION OF AN IMAGE FLUORESCENCE MICROSCOPY 22 WHAT IS FLUORESCENCE 22 FLUORESCENCE MOLECULES 23 FLUORESCENCE INSTRUMENTATION 24 WIDEFIELD FLUORESCENCE MICROSCOPY CONFOCAL LASER SCANNING MICROSCOPY 27 POINT SCANNER 27 SPINNING DISK MULTIPHOTON MICROSCOPY DETECTION SYSTEMS SPECIAL FLUORESCENCE MICROSCOPY APPLICATIONS DECONVOLUTION SUPER-RESOLUTION MICROSCOPY LITERATURE 38 3

4 INTRODUCTION Microscopy enables a direct imaging of organisms, tissues, cells, organelles, molecular assemblies and even individual proteins. Analytical techniques, as used in molecular biology, provide data set which represents cellular processes in an abstract way. Microscopy has been and is still an important, complementary technique to visualize the macro and/or microscopic structure and to assign structure to function and vice versa. Proper use of the microscopes, interpretation of images, and choices of specific applications demand an understanding of fundamental underlying principles. Discrimination of artifacts from structures and observation of finest details is only possible with knowledge of how images are physically formed. We will cover in this script the basis of light microscopy techniques and it should lead to the understanding of principle operation for different light microscopes, image formation and limits in resolution. In addition to this script, during the course some sample preparation procedures will be explained. Magnification versus resolution Microscopes form enlarged images of small objects. From this affirmation it seems that magnification is the principle aim of microscopy but, any level of magnification can always be produced by successive enlargements of images. However, when doing such a test it becomes clear that no further level of detail can be achieved and that a useful magnification exists. Therefore, the magnification is not the main goal of microscopy, the resolution of fine details or closely spaced points instead. The resolving power depends on the instrument and the imaging radiation (e. g.: light / electrons / X-rays). Each microscope will specify the smallest detail that the system can resolve (possibly with an ideal sample). Resolution refers to the detail revealed in the image of a given specimen. Resolution requires adequate instrumental resolving power but also depends on the contrast properties of microscope and specimen. Details can only be perceived unless they have a variation in intensity and are visible. The typical size of eukaryotic cells is micrometer and their subcellular components can not be seen by naked eyes. Additionally, cells are colorless and transparent. The organelles or other constituents have to be stained to make them visible in the light microscope. A variety of histological stains and fluorescent markers have been developed to this end. Some imaging techniques, their resolution limit, scales and corresponding objects are depicted in Figure 1. Figure 1: Resolution limit of different imaging techniques, involved radiation and size of biological object. 4

5 Imaging the ultrastructure of cellular components requires electron microscopes which have a much higher resolving power than light microscopes. The physical properties of electron microscopes demand specific preparation and staining techniques to reveal the ultrastructure of cells and tissues. FUNDAMENTALS OF LIGHT MICROSCOPY A compound light microscope is an optical instrument that uses visible light to produce a magnified image of an object that is projected onto the retina of the eye or onto an imaging device. The word compound refers to the fact that two lenses, the objective lens and the eyepiece (ocular), work together to produce the final magnification M of the image: M final = M objective M ocular Basis of a microscope - Types Figure 2 shows an image of a typical compound bright field (left) and fluorescence (right) microscope. For ease of operation objectives are mounted on a turret to allow changing of objectives. In addition, many different fluorescent filters can be inserted in a filter wheel. These type of microscopes allow imaging in different imaging modes like bright field, phase contrast, differential interference contrast or fluorescence microscopy depending on the demand. Figure 2: The compound light microscope. This illustration shows on the left an ordinary microscope without fluorescence illumination, on the right a fluorescence microscope. The illustration above the microscopes illustrates the two different illumination methods: transmitted versus fluorescent illumination. Figure 3: Observation of a magnified image of a specimen in the microscope. The objective lens forms a magnified image of the object. This intermediate image is observed with the eyepiece. The image obtained by the objective in the image plane of the microscope becomes further magnified for observation by eye. An ocular or eyepiece produces a virtual image that is seen by eye. This is illustrated in Figure 3 the virtual image perceived by eye seems to be located 5

6 about 25 cm in front of the eye. Please, note that an electronic camera directly records the real intermediate image formed by the objective lens. Apertures An aperture is a small opening to eliminate stray light from entering the light path. Fixed circular apertures are found at the rear or back focal plane of the objective. Variable apertures are built in the condenser and above the lamp the aperture diaphragm or field diaphragm. The exact location of the planes and apertures is adjusted for optimal imaging. This is achieved with Koehler illumination which is used in practically all techniques of light microscopy used in life science. Aperture and image planes Principles of geometrical optics show that a microscope has two sets of conjugate focal planes. Object or field planes are seen in normal viewing mode and contain all formed images, the object but also the illumination field diaphragm above the lamp. Aperture or diffraction planes (see below) can only be discerned by a special eyepiece telescope called a Bertrand lens which is focused on the back aperture of the objective lens. These planes are shown in Figure 4. The planes are called conjugate because the planes of a given set can all be seen simultaneously when looking in the microscope. Figure 4: The locations of conjugate focal planes in a light microscope adjusted for Koehler illumination. Koehler illumination Illumination is the most critical determinant of optical performance in light microscopy. It is important that light emitted from different locations on the filament be focused at the front aperture of the condenser. The size of the illuminated field at the specimen is adjusted so that it matches the specimen field diameter of the objective lens being employed. Because each source point contributes equally to illumination in the specimen plane, variations in intensity in the image are attributed to the object and not to inhomogeneous illumination. Koehler illumination fulfills these requirements. Under Koehler illumination a collector lens on the lamp housing is adjusted so that it focuses an image of the lamp filament at the front focal plane of the condenser while completely filling the aperture. Illumination of the specimen plane is bright and even. This is even achieved with irregular light sources such as a lamp filament. The precise positioning of the two different sets of conjugate planes is a strict requirement for maximal resolution and optimal image formation. This is achieved by: 1. Lamp focused on the front aperture of the condenser (located in diffraction planes use Bertrand lens). 6

7 2. Focus the specimen. 3. Focus the condenser to see the field stop diaphragm. 4. Adjust the condenser diaphragm (also phase rings) using the eyepiece telescope or Bertrand lens. The field diaphragm is open only until it is just not seen anymore in the image. Opening the field diaphragm too much will lead to stray light entering the light path. Objective designs and specifications Depending on the application different objectives with appropriate imaging performances are selected. As we will see below, the resolution is determined by the aperture. But resolution is only one of several points to consider. The type of immersion, contrast generation for transparent samples, working distance, flatness of field are others. All information about the objective (Figure 5) can be found on the objective barrel and should be carefully studied (and possibly written down in the lab journal) before any work is started. Figure 5. Markings on objectives: From top to bottom: company; type of objective; magnification aperture immersion medium; application or contrast formation; lens to image distance coverslip thickness (mm) working distance; color code for magnification code for immersion fluid. Achromats are blue red corrected (486 nm and 656 nm) and spherical correction is done at yellow-green (540 nm). These objectives are satisfactory with white light and excellent using monochromatic light up to medium magnifications. Apochromats are highly color corrected designs suitable for color recording using white light or multi-color fluorescence. These lenses are red, green, blue and dark blue corrected for color and corrected for spherical 7

8 aberration at green and blue wavelengths. In plan apochromats the curvature of field is also corrected. New designs are transparent and corrected to near UV light or optimized for infrared imaging in multiphoton imaging. These lenses have the highest NA (up or even above 1.4) and are the prime lenses for fluorescence microscopy and low light level application. Objectives and imaging errors Aberration of lenses Simple lenses have spherical surface which have many intrinsic optical faults called aberrations that distort images. Major faults are: Chromatic aberration Spherical aberration Figure 6. Chromatic aberration occurs because a lens refracts light differently depending on the wavelength. Blue light is bent inward toward the optic axis more than red light. Blue wavelengths are focused in an image plane closer to the lens than the image plane for red wavelengths. The result is disastrous: point sources are surrounded at best by color halos and colors will change depending on the focus of the objective with an image never becoming sharp. This fault will be corrected by using glass with different color-dispersing properties (e. g. achromatic lens). Figure 7. Spherical aberration occurs on spherical surfaces where central and peripheral rays are focused at different axial locations. Again, there is not a well defined image plane and a point source of light at best focus appears as a spot surrounded by a bright halo. An extended object will be blurred. The correction will be done by employing not only one lens but a combination of positive and negatively curved surfaces. Important to note is that the coverslip used to cover the sample is part of the optical system and must be of correct thickness (number 1.5 or 170 µm). Some objectives contain correction collars for adjustment of spherical aberration due to refractive index mismatch or coverslip thickness changes. Coma Figure 8. Coma refers to a streak of light. Rays from points which are off axis are focused closer to the optical axis when going trough the periphery of the lens than when going through the center. 8

9 Astigmatism Figure 7. Astigmatism is an off axis aberration. Rays going trough the horizontal and vertical diameters of a lens are focused as a short streak at two different focal planes. The streaks appear as ellipses drawn out in horizontal and vertical directions. Curvature of field Figure 9. Curvature of field is also an off axis aberration. It appears that the image plane is not flat but has a concave spherical surface as seen from the objective. It is important to note, that an image is required to be flat for image recording. Distortion of field Field distortion is an aberration that causes the focus positions of the object image to shift laterally in the image plane with increasing displacement of the object from the optic axis. The correction for these faults is done using glass elements with different refractive indexes, color dispersion, incorporation of aspherical lens surfaces to name the most important. Faults are corrected but never completely removed. It is a compromise to correct some errors while other may even worsen. For these reasons, objective lenses vary considerably in design, optical performance and cost. Lamps and specimen illumination Optimal imaging performance in light microscopy is achieved by properly illuminating the specimen. Apart from Koehler illumination the proper selection of wavelength, intensity and alignment of the lamp are a must. A research microscope may be equipped with a variety of different illuminators but also filters to attenuate the light intensity and wavelength. 9

10 Incandescent filament lamps Frequently used for transmission illumination are tungsten or halogen lamps. Both are convenient, inexpensive and provide bright, even illumination when used with a ground glass filter. Both types of lamp produce a continuous spectrum of light across the visual range with a peak output in the red and infrared (Figure 10). Figure10: Spectra of various lamps. Tungsten and halogen lamps are primarily used for transmission imaging (bright field, phase contrast, differential interference contrast, polarization). Hg (mercury) and Xe (xenon) arc discharge bulbs provide x more light and are used for fluorescence microscopy. The light emission is regulated via a variable power supply. It is important to note that by changing the power of a lamp its spectrum is shifted. When producing color images (e. g. in histology) a specific voltage should be selected to obtain consistent and reproducible spectrum of wavelength. The effect of changing the voltage is illustrated in Figure 11. Figure 11: a) Quartz halogen tungsten filament lamp. By changing the voltage of the lamp the emission spectrum is shifted to the red corresponding to lower visible wavelength. b) Image on the left has been acquired with correct voltage. Image on the right was acquired with a low voltage on the lamp to reduce light intensity. As a result of the spectrum shift, the image has a red tint. Ion arc lamps Ion (mercury or xenon) arc lamps are 100x to 1000x times brighter compared to tungsten filament lamps. When combined with appropriate filters (see fluorescence) they can provide brilliant monochromatic illumination. The increase in brightness comes with several inconveniences: short lifetime (200 hours) and higher costs (300 CHF). Arc lamps need expensive stabilized power supplies and should not be turned on when still hot. It is advisable to leave these lamps on even if there is a pause of 60 minutes. The emission spectra of these arc discharge lamps are shown in Figure 10. Metal halide lamps Metal halide lamps are becoming popular because they are bright (150 W), have a long lifetime (1000 hours). In contrast to arc discharge lamps they do not need complicated adjustment and the light can be delivered to the microscope via a light guide. This has the additional benefit of separating a considerable heat source away from the microscope. 10

11 Lasers Laser light sources are primarily used in all different types of laser scanning microscopes (confocal laser scanning, multiphoton) but also in some specialized applications like total internal reflection fluorescence microscopy (TIRF). Lasers produce their light by stimulated emission of light. This process leads to monochromatic light which is coherent. Costs compared to other light sources are 100x to 1000x higher. As lasers are monochromatic they are ideally suited for applications involving fluorescence but with the drawback that for excitation of different fluorochromes also different lasers must be employed. This leads not only to very high costs but also to fairly complicated systems. IMAGE FORMATION BY LENSES AND LIGHT As mentioned above, light microscopes use light to illuminate an object and to generate, via objectives, an image of the object. This has not been changed since a few hundred years since its invention. However, advances in science and technology have profoundly changed light microscopy over the past ten to twenty years. Nevertheless, physics behind image formation is identical but, detectors and light sources have changed dramatically allowing these advances. How is an image formed in a microscope Two microscope components are of critical importance in forming the image: 1. The objective lens, which collects light that has interacted with the specimen and forms a magnified real image at the intermediate image plane. 2. The condenser lens, which focuses light from the illuminator onto a small area of the specimen. Light as a probe of matter It is useful to think of light as a probe that can be used to determine the structure of objects viewed under the microscope. Generally, the probes must have size dimensions that are similar to or smaller than the structures (sample) being examined (note: compare the probe size (electrons) and resolution of electron microscopes). As an approximation, the resolution limit of the light microscope with an oil immersion objective is about onehalf of the wavelength of the light employed. Knowledge of the properties of light is important in selecting filters and objectives, interpreting colors, performing low-light imaging, and many other tasks. Light can also be described as a particle or a wave. The wavelength corresponds to the energy of light, the amplitude to intensity. The wave like property of light is specifically the electric and magnetic field which fluctuates over time (Figure 12: Light as an electromagnetic wave. The electric field (E) vector is shown to oscillate vertically, the magnetic field (B) vector horizontally. Figure 12: Light as an electromagnetic wave. The electric field (E) vector is shown to oscillate vertically, the magnetic field (B) vector horizontally. Light interacts with matter in a variety of ways. Light incident on an object might be absorbed, transmitted, reflected, or diffracted. Such objects are opaque, transparent, reflective, or scattering. Light may be absorbed and then re-emitted as visible light or as heat. Objects or molecules that absorb light transiently and quickly re- 11

12 emit it as longer wavelength light are described as being phosphorescent or fluorescent depending on the time required for re-emission. Light can be scattered by small particles and structures having dimensions similar to the wavelength of the light itself. The diffraction of light by small structural elements in a specimen is the principal process governing image formation in the light microscope. In Figure 13 are represented some properties of light which often play a role in microscopy. Monochromatic light is composed of only one wavelength. In coherent light all electromagnetic field vectors are in oscillating in phase. Laser light is monochromatic, but also coherent. Linear polarized means that all electric fields of individual waves are in one plane. White light is composed of many different wavelength (polychromatic), is non coherent and not polarized. Figure 13: As an analytic probe used in light microscopy, the kind or quality of light is described according to the degree of uniformity of rays comprising an illuminating beam. The kinds of light most frequently referred in microscopy are depicted. Image: Olympusmicro.com Diffraction and image formation Diffraction and interference are the key principles that determine how a microscope forms an image. It is important to note and mentioned here again: objectives do not just produce a magnified image of an object. In transmitted light the illuminating beam is diffracted or scattered by the specimen, collected by the objective lens, and focused in the image plane, where waves constructively and destructively interfere to form a contrast image. The scattering of light (diffraction) and its recombination (interference) are phenomena of physical optics or wave optics. In self luminous objects or fluorescent objects waves radiate in all direction. A part of these waves is collected by the objective and similar as described above, an image is formed by recombination in the image plane. In Figure 14 an example of diffraction and image formation by interference is shown. 12

13 Figure14: Top: Diffraction and image formation by a grating. A plane wave is diffracted at the grating and a magnified image is produced on the image plane. In the back focal plane, a diffraction pattern is produced by interference of rays that have the same path length. Bottom: Real diffraction pattern at the back focal plane inside an objective. Note that the diffraction pattern is inversely proportional to the grating and only a certain number of diffraction spots are collected by the objective. Light that is diffracted and not collected by the objective does not contribute to the diffraction pattern and hence, to the image formed in the image plane (see Figure 15). Therefore, the larger the angle of light an objective can collect the better the resolution. Figure 15: Generation of an image by interference requires collection of two adjacent orders of diffracted light by the objective lens. In a and b no diffracted light is collected and no image will be formed. In c a minimum of diffracted light is collected (1 st order diffraction) and an image is formed. Higher order diffractions collected in d will lead to a higher definition of the image. The image of a point source of light is a diffraction pattern created by the action of interference in the image plane. When highly magnified, the pattern is observed to consist of a central spot or diffraction disk surrounded by a series of diffraction rings. The central diffraction spot (Figure 16) is also called the Airy disk (Sir George Airy, ). 13

14 Figure16: The intensity distribution of point sources. Airy patterns are formed from single points. High numerical aperture objectives capture more of the diffracted orders and produce smaller size disks than do low numerical aperture objectives. The larger disk sizes in (a) and (b) are produced by objectives with lower numerical aperture, while the very sharp Airy disk in (c) is produced by an objective of very high numerical aperture. Figure source: Zeiss-campus.magnet.fsu.edu Resolution The key element in the microscope s imaging system is the objective lens, which collects the diffracted ray and forms by interference of diffracted and non diffracted rays an image in the image plane. The collection of diffracted rays is directly proportional to the numerical aperture, the angle over which the objective can collect diffracted rays from the specimen. The numerical aperture is the key parameter determining spatial resolution. The greater the number of higher diffracted orders admitted into the objective, the smaller the details of the specimen that can be clearly resolved. The numerical aperture (NA) is described as: NA = n sin(α) α is the half angle of the cone of specimen light accepted by the objective lens and n is the refractive index of the medium between the lens and the specimen. For dry lenses used in air, n = 1; for oil immersion objectives, n = The diffraction angles capable of being accepted by dry and oil immersion objective lenses are compared in Figure 17. By increasing the refractive index of the medium between the lens and coverslip, the angle of diffracted rays collected by the objective is increased. Because immersion oil has the same refractive index as the glass coverslip, refraction of specimen rays at the coverslip air interface is eliminated, the effective half angle is increased, and resolution improved. Figure 17: Light path of light reflected or emitted with a oil immersion (right) versus a air objective. Due to total reflection of rays under a angle larger than the critical angle only light from a smaller angle is accepted by a lens operated in air. Figure source: Zeiss-campus.magnet.fsu.edu 14

15 Resolution of a point object Rayleigh criterium For point objects that are self-luminous or for nonluminous points that are examined by bright-field microscopy 1 the resolving power of the microscope is defined as: d = 0.61 λ/na d is the minimum resolved distance in µm, λ is the wavelength, and NA is the numerical aperture of the objective lens. The equation describes the Rayleigh criterion for the resolution of two closely spaced diffraction spots in the image plane. By this criterion, two adjacent object points are defined as being resolved when the central diffraction spot (Airy disk) of one point coincides with the first diffraction minimum of the other point in the image plane. This is the Rayleigh criterion for spatial resolution. Figure18: The intensity distribution of point sources. In d, two points are overlapping. In e, according to the Rayleigh criterium, the two points are just resolved (note that dark regions correspond to high light intensity, and vice versa). An image of an extended object consist of a pattern of overlapping diffraction spots, the location of every point x, y in the object corresponding to the center of a diffraction spot x, y in the image. An object can be regarded as composed of submicroscopic particles each and independently diffracting light. For a given objective magnification, changing the objective for one with the same magnification but a higher NA, the diffraction spots in the image grow smaller and the image is better resolved. Thus, larger aperture angles allow diffracted rays to be included in the objective, permitting resolution of specimen detail that otherwise might not be resolved. The depth of field and depth of focus Just as diffraction and the wave nature of light determine that image of a point object is a diffraction disk of finite diameter, so do the same laws determine that the disk has a measurable thickness along the z-axis. Depth of field in the object plane refers to the thickness of the optical section along the z-axis within which objects in the specimen are in focus. For diffraction-limited optics, the wave-optical value of dz is given as: dz = n λ/na 2 Where n is the refractive index, λ the wavelength and NA the numerical aperture. Depth of focus is similar to depth of field but is measured in the image plane. Both terms are often used interchangeably. Examples of depth of field calculations but also images of a bead are shown in Figure 19 and Figure Condensor NA > objective NA 15

16 10.0 dz (um) Figure 19: Depth of field of a widefield microscope in relation to the numerical aperture NA Figure 20: Depth of field of a self luminous point source (bead) acquired using a confocal laser scanning microscope. Note the non isotropic form of the cross section due to spherical aberration and the lower resolution in Z. Image acquired with an oil immersion objective, 63x, NA 1.4 PHASE CONTRAST MICROSCOPY AND DARK FIELD MICROSCOPY In stained histological specimens, when illuminated with white light, specific wavelengths are absorbed by dyes allowing objects to appear in color. These are amplitude objects. However, unstained objects, such as cells, are essentially invisible in an ordinary bright field microscope. They do not adsorb light but, they slightly alter the phase of the light diffracted by the specimen, usually retarding such light approximately ¼ wavelength as compared to the undeviated direct light passing through or around the specimen unaffected. They are called phase objects. Our eyes as well as cameras are unable to detect these phase differences. Figure 21: Effects of amplitude and phase objects on the waveform of light. Top: reference ray with specific amplitude, wavelength and phase. Middle: a pure amplitude object absorbs energy and reduces the amplitude but does not alter the phase or wavelength. Bottom: a pure phase object alters velocity and shifts the phase but amplitude and wavelength are not changed. 16

17 Phase contrast microscopy Their images have not enough contrast because almost no light is absorbed by the specimen. However these objects not only diffract light but, also induce, upon interaction, a phase shift. In phase contrast microscopy the phase shift which cannot be seen by eye is transformed into an amplitude difference that can be seen. In phase contrast microscopy the aim is to isolate the surround and diffracted rays emerging from a specimen so that they occupy different locations in the diffraction plane at the back aperture of the objective lens. Subsequently, the phase is advanced and its amplitude reduced of the surround light. This maximizes the differences in amplitude between the object and background in the image plane. To isolate the surround and diffracted rays the sample is illuminated in a special way. A condenser annulus (transparent ring) in the front aperture of the condenser illuminates the specimen in a kind of hollow cone of mostly parallel light with a dark center. Figure 22: Principle of phase contrast microscopy. The specimen is illuminated through a condenser annulus. The image of the condenser annulus is in the back focal plane of the objective again in a ring like structure. Diffracted (and phase shifted) rays are found in the rest of the back focal plane. A phase plate reduces amplitude and advances the phase of non diffracted light in the back focal plane. Light that does not interact with the specimen are focused as a bright ring of 0 th order diffraction in the back focal plane of the objective (diffraction plane conjugate to the condenser front aperture with the condenser annulus!). Light that is diffracted by the specimen traverses the diffraction plane at various locations across the entire back aperture. The exact location depends on the local refractive index in the scattering objects in the specimen. With this setup non diffracted and diffracted light is spatially separated in the diffraction plane and can be manipulated. The non diffracted light or surround wave (S wave) is advance in phase by + λ / 4 and reduced in amplitude by approximately 75%. Diffracted waves going through the thicker part of the phase plate are retarded in phase by approximately - λ / 4. Rays will proceed to from an image in the image plane by interference. Due to the change in phase in the phase plate of diffracted and non diffracted waves form an image by interference where phase differences are transformed into amplitude differences. It is important to realize that phase contrast microscopy translates optical path length and refractive index both into differences in light intensity, thus providing contrast. Objects appear on a dark grey background surrounded by halos. Objects having the same contrast in phase contrast microscopy can vary both in size and refractive index. Large homogeneous object which do not show any difference in path length and refractive index will not show any contrast difference in phase contrast. 17

18 S wave D wave P wave S wave D wave P wave Amplitude Amplitude a) nm b) nm Figure 23: Phase relation of diffracted waves (D wave), non diffracted (S wave surround wave) and particle wave (P wave). a) in bright field microscopy the amplitude difference between the particle (S+D) and surround wave (S) is too small to be detected. b) In phase contrast microscopy the phase of the surround wave is shifted by λ/4 and its amplitude by 75%. The diffracted wave is additionally retarded by λ/4 resulting in a negative interference resulting in a particle wave that can be discriminated from the background (S wave). Dark field microscopy In most forms of transmitted light microscopy, both the diffracted rays (rays that interact with the specimen) and non diffracted rays (rays that pass undeviated through the specimen) are collected by the objective lens and contribute to the image formation. One solution to provide contrast in transparent objects, where the fraction of non interacting rays is very high, is dark field microscopy. The technique is based on small amounts of diffracted light from small phase objects. Objects appear on a dark, black background, because no direct light is allowed to enter the objective lens. Dark field conditions are obtained by illuminating the specimen at an oblique angle such that direct, nondiffracted rays are not collected by the objective lens. This is obtained by using a special dark field condenser annulus, which is mounted in the condenser turret. Special oil immersion dark field condensers must be used for oil immersion objectives. Figure 24: Optical scheme for dark field microscopy. The geometry allows only diffracted light to be collected by the objective lens. Direct, nondiffracted light are inclined at a steep angle and miss the objective entirely. The interpretation of dark field images is similar to self luminous or fluorescent objects on a dark background. However, edges of extended, highly refractile objects diffract the greatest amount of light and dominate the image. Important for high definition dark field images is the use of high NA objectives. The aperture should not restrict too many of the diffracted waves which would result in a loss of definition of fine details in the specimen. Dark field imaging is advantageous for imaging small objects like bacteria, diatoms, but even lysosomes or microtubules can be discerned. Dark field optics is inexpensive, simple to employ and does not need any special equipment like special objectives, illuminators or prisms and filters. DIFFERENTIAL INTERFERENCE CONTRAST MICROSCOPY Differential interference contrast microscopy (DIC) requires plane-polarized light and additional light-shearing (Nomarski) prisms to exaggerate minute differences in specimen thickness gradients and refractive index. Lipid bilayers, for example, produce excellent contrast in DIC because of the difference in refractive index between aqueous and lipid phases of the cell. Differential interference contrast is an important technique for imaging thick animal tissues because, in addition to the increased contrast, DIC exhibits decreased depth of field at wide apertures, creating a thin optical section of the thick specimen. 18

19 In the previous chapter phase contrast optics was used to convert an optical path difference from refractive index changes into a contrast difference in the object image. However gradients in optical path lengths can also be detected using differential interference contrast (DIC) and modulation contrast (not covered here) microscopy. The difference of these two techniques to phase contrast is that images do not contain phase halos; it is possible to optically section through a sample. a) b) Figure 25: Comparison of phase contrast and DIC image of HeLa cells in culture. a) phase contrast and b) DIC image. Due to the very thin thickness of HeLa cells in culture the halo around the cells is not very prominent. DIC images appear deceptively three dimensional but contrast reflects changes in optical path length and refractive index. DIC components DIC microscopes are dual beam interference systems that transform local gradients in optical path length (refractive index, thickness) into contrast. The specimen is sampled by pairs of closely spaced rays that are generated by a beam splitter. If the pair of rays encounter regions with different refractive index both rays will have a different phase due to the different path length and this different phase of the two rays is transformed into contrast. To differentiate two rays polarized light with mutually distinct plane of polarization is employed. A number of optical components is used for DIC and will be explained below. Polarizers Light from most illuminators (exception are lasers) is nonpolarized, the E (see Figure 12, page 11) of all waves vibrating in all possible angles with respect to the axis of propagation. In a ray of linearly polarized light, the E vectors of all waves vibrate in the same plane. A device or filter that produces polarized light is called a polarizer. When the same filter is used to detect the plane of vibration it is called an analyzer. Efficient polarizers are made of transparent crystals such as calcite or using partially light absorbing sheet of linear polarizing material (e. g. organic polymers of Polaroid filters). A polarizer in front of the condenser produces polarized light. The plane of vibration of the E vector of the waves is oriented horizontally on an east west or right left line when looking into the microscope. Wollaston prisms or DIC prisms A condenser DIC prism is used close to the front aperture of the condenser to act as a beam splitter. Every incident ray of polarized light entering the prism is split into two rays that function as the dual beams of the interference system. After the objective a similar prism, the objective DIC prism used to recombine the beams for analysis of phase differences. To split polarized light into two beams birefringent or double refracting crystals are used. This property is found in many crystals and minerals such as quartz, calcite or tourmaline. These crystals are optically anisotropic because atoms in these crystals are ordered in a precise geometric arrangement causing direction dependent differences in the refractive index. In contrast, a sheet of glass is an amorphous mixture of silicates and is isotropic and not double refractive. When a ray of light is incident on a birefringent crystal it becomes split into two ray that follow separate paths (Figure 26). One ray, the ordinary ray or O ray, observes the laws 19

20 of normal refraction, while the other ray, the extraordinary ray or E ray, travels along a different path. Therefore, for each ray entering a birefringent crystal a pair of mutually perpendicular polarized O and E rays emerge. Figure 26: Splitting of an incident ray into O and E ray by a birefringent crystal. The optical axes are oriented differently in the left and right crystal as shown by the double arrow. In the left crystal a ray incident at an angle oblique to the optic axis of the crystal is split into O (dashes) and E (dots) rays. The plane of polarization is mutually perpendicular and the both rays have optical path difference (are out of phase after passing the crystal). If the propagation axis of a ray is perpendicular to the optic axis it will be split into O and E rays that follow the same trajectory but have an optical path difference To split or shear (at the condenser with the condenser prism) the polarized light into two mutually perpendicular polarized beams a Wollaston prism is used (Figure 27). In a standard Wollaston prism the optic axes of two cemented wedges of calcite or quartz are oriented parallel to the outer surfaces of the prism and perpendicular to each other. The shear axis and the separation distance between the resultant O and E rays are the same for all O and E ray pairs across the face of the prism. A similar Wollaston prism is also used to recombine the two rays after the objective (objective Wollaston prism). In some microscope designs (e. g. Leica) modified Wollaston prisms (also called Nomarski prism) are used. One of the axes is not perpendicular to the other in the wedges with the result that the apparent crossing of the two beams seems to be outside the wedges and not inside the wedges. This has some advantages when used as an object prism. A standard Wollaston prism has to be located close to the back focal plane of an objective which is inside the objective barrel in high power objectives. A Nomarski or modified Wollaston prism can be placed several millimeters away from the back focal plane allowing more freedom in the microscope design and is also advantageous in optical performance. It is important to note that the condenser and objective Wollaston prism must be matched to the objective used. The amount of shear induced must be in the order of the resolution of the objective. Figure 27: Design and action of a Wollaston prism. Wollaston prisms are used in DIC microscopy to generate and recombine pairs of O and E rays. Shown is the action of a condenser prism. The optic axes of the wedges are perpendicular on each other and the two beams separate (or recombine in the objective prism) inside the condenser prism. Analyzer An analyzer (polarization filter) is used to analyze rays of plane and elliptically polarized light coming from the objective and to transmit plane polarized light that is able to interfere and generate an image in the image plane. Its vibrational plane is oriented vertically in a north south or top bottom orientation when facing and looking into the microscope. Setup of a DIC microscope and formation of an image Apart from the above mentioned components special strain free lenses are highly desirable, because ordinary objectives contain stress signatures in the glass from pressure points in the lens mounting and inhomogeneities in the glass. This would decrease contrast. 20

21 Figure 28: Optical components of a DIC microscope. Two polarizers and two modified Wollaston prisms are required. The condenser DIC prism acts as a beam splitter, producing two closely spaced parallel beams that traverse the object and are recombined by the objective DIC prism. Davidson and Abramowitz Optical microscopy. Ray tracing and wave optics are both useful for understanding image formation in a DIC microscope (Figure 29). The trajectories of rays from polarizer to image plane explain the action of optical components and make it clear that the DIC microscope is a double beam interference system. 1. An incident ray of linear polarized light is split by the condenser prism into a pair of O and E rays separated by a small distance (below or at the resolution of the objective). The plane of polarization of O and E rays is perpendicular to each other 2. The O and E rays remain parallel between condenser and objective. Thus every point in the specimen is sampled by pairs of beams that provide dual beam interference in the image plane. There is no universal reference wave as in some other interference systems. 3. The O and E rays are recombined after the objective by the objective DIC prism. 4. An analyzer transmits light if an elliptically polarized ray is generated by the recombination of an O and E ray encountered a phase difference. Figure 29: Progression of rays through the DIC microscope. An incident beam of linearly polarized light is split by the condenser DIC prism into O and E rays that are focused by the condenser lens onto the specimen. In the absence of an optical path difference (left and middle situation), the O and E rays are combined by the objective prism, giving linearly polarized light that vibrates in the same plane as the polarizer and is completely blocked by the analyzer. Only with an optical path difference with phase shift (situation right) the objective prism recombines the beams giving elliptically polarized light that is partially transmitted by the analyzer. 21

22 Fluorescence microscopy FLUORESCENCE MICROSCOPY With light microscope optics adjusted for fluorescence microscopy, it is possible to examine the distribution of a single molecular species in a specimen, and under special conditions, even detect individual fluorescent molecules. In contrast to other forms of light microscopy based on object-dependent properties of light absorption, optical path differences, and phase gradients, fluorescent microscopy allows visualization of specific molecules that fluoresce in the presence of excitatory light. Thus, the amount, intracellular location, and movement of macromolecules, small metabolites, and ions can be studied using this technique. Figure 30. Synaptic components can be resolved by fluorescence microscopy. (A) Postsynaptic element labelled with Calbindin (Horizon V-500) and (C) presynaptic boutons stained with VGluT 1 (Oregon green). Overlay of two images showing the close apposition of both structures. Scale bars: A-C 30 and D-F 3 microns. What is fluorescence Fluorescence is the emission of photons by atoms or molecules whose electrons are transiently stimulated to a higher excitation state by radiant energy from an outside source the excitation light (see Jablonski diagram, Figure 31). Relaxation of electrons to the ground state is relatively fast in 10-9 to s with release of energy as a photon. Since some energy is lost in the process, the emitted fluorescent photon typically exhibits a lower frequency of vibration and a longer wavelength than the excitatory photon that was absorbed (Stoke s shift). Concurrently other processes may occur like intersystem crossing to triplet states with subsequent relaxation via phosphorescence or internal conversion without emission of radiation. The triplet state is important because molecules in the triplet state are reactive, which in turn can lead to the production of free radicals and subsequent photobleaching. 22

23 Fluorescence microscopy Figure 31: Jablonski diagram showing energy levels occupied by an excited electron within a fluorescent molecule. Collapse to the ground state after excitation can occur through any of the following three pathways: fluorescence emission, intersystem crossing and generation triplets, internal conversion and phosphorescence. Fluorescence molecules Dyes Molecules that are capable of fluorescing are called fluorescent molecules, dyes or fluorochromes. If a fluorochrome is conjugated to a larger macromolecule, the tagged macromolecule is said to contain a fluorophore, the chemical moiety capable of producing fluorescence. Fluorochromes exhibit distinct excitation and emission spectra that depend on their atomic structure and electron resonance properties. Figure 32: Left. Absorption (excitation) and emission spectra are normally slightly overlapping but the emission spectra is shifted (Stoke s shift) to longer wavelength. Before starting a labelling experiment with different fluorochromes, check the specifications of the fluorochromes and select a combination with well separated excitation and emission spectra. Top. Fluorescein absorption and emission spectra. The absorption is always at lower wavelengths. The difference between the absorption and emission spectra is called the Stokes shift. There is a large repertoire of dyes and fluorochromes available. Differences are based on the absorption / emission properties, the quantum efficiency ( how much fluorescence is generated or more specifically the fraction of absorbed photon quanta that is reemitted by a fluorochrome) and the resistance to photobleaching. 23

24 Fluorescence microscopy Fluorescent dyes are often coupled to antibodies and can be used mainly in fixed cells for staining of specific molecules. Finally, the shapes of spectral curves and the peak wavelengths of absorption and emission spectra vary, depending on factors contributing to the chemical environment of the system, including ph, ionic strength, solvent polarity, O 2 concentration, presence of quenching molecules and others. These factors explain why the fluorescence of a dye such as fluorescein varies depending on whether it is free in solution or conjugated to a protein or other macromolecule. When selecting fluorescent dyes important parameters are a large Stokes shift (20nm for fluorescein), quantum efficiency, molar extinction coefficient (potential to absorb photons) but also resistance to photobleaching, solubility and chemical stability. New dyes such Alexa dyes or cyanine dyes have much higher quantum yields and stability and should be used whenever possible. Information about dyes, fluorescent molecules to label organelles or are even ion specific can be found on the web (Invitrogen: Jackson ImmunoResearch Laboratories: Fluorescent proteins Green fluorescent protein (GFP) isolated from the jellyfish Aequorea victoria and its mutated forms, blue, cyan, yellow are used to produce fluorescent chimeric proteins that can be expressed in living cells, tissues and whole organisms. Recently, many other fluorescent proteins were found, mainly in marine organisms, and cloned. Examples are DsRed, mcherry or RFP, all in the red. New developments try to modify fluorescent proteins to have variants that emit light in the blue. Current blue or blue shifted variants tend to have severe disadvantages for work like low stability and quantum yield. Fluorescent proteins as genetically encoded fluorescent tags can be used in a large number of experiments. They can be cloned to proteins for localization or expression studies, used as read out in gene expression or as reporters studies to name only a few. It is important to note that all fluorescent proteins, as genetically encoded probes, need time to be expressed and for the autocatalytic formation of the fluorophore, that is composed of amino acids. The molecular structure of GFP is shown in Figure 33. a) b) c) Figure 33: Structure of GFP. a) Side view and b) top view of a monomer composed of a central -helix surrounded by an eleven stranded cylinder of anti-parallel -sheets. The cylinder has a diameter of about 30Å and is about 40Å long. The fluorophore (c,) yellow) is located on the central helix inside the cylinder where it is protected by the very stable β-can barrel structure. It is common practice to label cells with multiple fluorescent dyes to examine different molecules, organelles or cells in the same preparation using different fluorescence filter sets. How the system has to be set up for fluorescence work also involving multiple fluorescent dyes is shown below. Fluorescence instrumentation Detection of fluorescent dyes requires proper illumination of the sample, namely precise excitation and detection of emission but also removal of unwanted reflected illumination light. This is classically done using filter as shown in Figure

25 Fluorescence microscopy The illumination is done with ion arc lamps (mercury, xenon) or metal halide lamps because only a narrow band of wavelengths, and consequently a small portion of the illuminator output, is used to excite fluorochromes in the specimen. For efficient high contrast imaging, both the illuminator and objective lens are positioned on the same side of the specimen. In this arrangement, the lamp and light delivery assembly are called an epi-illuminator, and the objective lens functions both as the condenser and objective. Fluorescence filter sets are used that contain three essential filters: excitation filter, dichroic mirror, barrier (or emission) filters. These filters are positioned in the optical path between the epi-illuminator and the objective. High NA immersion objectives are used. These objectives are made of low-fluorescence glass and the high NA maximizes light collection to provide maximal resolution and contrast. Figure 34: Arrangement of filters in a fluorescence filter cube. The diagram shows the orientation of filters in a filter cube in an epi-illuminator. The excitation beam (dotted line) passes through the exciter and is reflected by the dichroic mirror and directed toward the specimen. The return beam of emitted fluorescence wavelengths (solid line) passes through the dichroic mirror and the emission filter. Excitation wavelengths reflected at the specimen are reflected by the dichroic mirror back towards the light source. Excitation wavelengths that manage to pass through the dichroic mirror are blocked by the barrier (emission filter). Filters Filters (see Figure 35 for the optical setup in the microscope) used to isolate bands of wavelengths in fluorescence microscopy include coloured glass filters, thin film interference filters or any combinations of these filter types. The filter includes long-pass, short-pass, narrow or broad bandpass filters. The action of these filters is shown in Figure 34. Figure 35: Function of different filter types. Short pass are transparent for wavelength lower than the 50% cut-off wavelength. Long pass are transparent above the 50% cut-off wavelength. Band pass filters allow one (or several) wavelength band to pass, but block wavelength outside the transmission band. They are characterized by the halfbandwith (HBW) 25

26 Fluorescence microscopy or full width at half maximum (FWHM) and the central wavelength (CWL). Dichroic mirrors Dichroic mirrors or beam splitter are special long pass filters coated with multiple layers of dielectric material optimized for reflection and transmission at certain boundary wavelengths. The mirror is mounted at a 45 angle with respect to the optic axis within a filter cube. It faces the light source and reflects the short excitation wavelengths at a 90 angle, but transmits long fluorescence wavelengths that are collected by the objective to the image plane. Depending on the application one can use dichroic mirror (but also excitation and emission filters) with multiple transmission peaks and troughs that correspond to the used fluorochromes (e. g. Figure 36) a) b) c) d) Figure36: a) Spectrum of DAPI, b) spectrum of FITC, c) spectrum of Alexa 594. d) Transmission profiles of a triple band filter set for DAPI, FITC and Texas red. Each of the three filters (left: excitation filter, middle: dichroic mirror, left: barrier or emission filter) contains multiple bandwidths that transmit or reflect three distinct bands of wavelengths simultaneously. The grey area designates the spectral region of the filter for FITC. Widefield fluorescence microscopy In wide field microscopy the whole sample is illuminated with light using epi-illumination and the appropriate filters to excite fluorescent dyes. Fluorescent imaging allows the localization of numerous molecules in high resolution by combining with antibody labelling, fluorescent protein expression systems or various dyes. In addition is also possible to investigate physiological processes like Ca 2+ concentration. Wide field microscopy is especially well suited to image living specimens under physiological conditions due to a relatively low phototoxicity. Acquisition of images can be in as fast as frames per second, but also low light imaging is possible with integration times of minutes to detect the faintest signals. The setup of a microscope for fluorescence imaging is shown in Figure

27 Fluorescence microscopy Figure 37: Comparison of a widefield microscope setup for bright field, phase contrast or DIC imaging (left) to a setup for widefield fluorescence imaging (right). The main difference is the epi-illumination in fluorescence imaging to increase quality of fluorescence detection. CONFOCAL LASER SCANNING MICROSCOPY In confocal laser scanning microscopy (CLSM) the sample is illuminated with laser light focused to a diffraction limited spot. The spot is scanned over the sample to excite fluorescent dyes. Emission of fluorescent light is detected through a pinhole located in a confocal plane relative to the plane of focus. Stray light from structures out of focus is blocked by the pinhole leading to well defined images not only in the imaging plane but also in axial direction (Figure 39). Variation of confocal laser scanning techniques exist which scan multiple points (spinning disk systems) or lines (line scanning) over the sample to increase speed in acquisition. Image acquisition speed of confocal laser scanning systems can be in the range of 25 images per seconds while simultaneous four channels can be detected in parallel. Data acquired using confocal laser scanning microscopes have a superior z-resolution and data can be acquired in three dimensions. Using environmental chambers even multicolor life imaging in three dimensions is possible. The comparison between widefield and laser scanning microscopes is shown in Figure

28 Fluorescence microscopy Figure 38: Comparison of a confocal laser scanning microscope (left) with a widefield microscope (right). Shown is a point scanning laser scanning microscope. Modern CLSM are integrated systems consisting of a widefield fluorescence microscope with multiple laser light sources, a confocal scan head with optical (mirrors, dichroic mirrors, filters, detectors), a computer and monitor for display. Software for acquiring, processing and analyzing images are included. Mirrors Mirrors in a CLSM are in most systems galvanometer based. For fast scanning systems (> 1600 Hz line speed) resonant scanners are used. Advantageous of using galvanometer based mirrors is the freedom to scan free forms or even lines. One mirror oscillates left right while the other moves up and down. Fluorescent light follows the same light path on its return and is brought to the same position on the optic axis as the original exciting laser beam. This process is called descanning. Filter and wavelength detection Excitation, dichroic mirrors and emission filters are similar to widefield microscope. Even when using monochromatic lasers, filter systems are needed, because in many cases multiline lasers or several lasers are used to excite multiple fluorochromes. New developments are using acousto-optical devices that can be electronically driven to reflect or transmit certain wavelength thereby making dichroic mirrors and excitation filters obsolete. Emission filters are more and more replaced by gratings or prism which split the emitted fluorescent light to separate detectors. 28

29 Fluorescence microscopy. Figure 39: a) Optical pathway in a confocal scan head. A laser beam is reflected by a dichroic mirror onto components of the scan control mechanism that sweeps the beam in a raster back and forth across the specimen. The objective lens collects fluorescent light, which is descanned at the scan control. b) Fluorescence from the diffraction limited excitation spot is projected to the image plane where the pinhole is located (conjugated or confocal plane). Out of focus fluorescence (generated by scattering of excitation or emission light) is blocked by the pinhole and is not detected Pinholes One or more variable pinholes are used for generating confocal images. The dimension of the pinhole determines the axial (and to a lesser decree the lateral) resolution (Figure 40). Normally the diameter of the pinhole is measured in number of Airy disks (see Figure 16). Opening the variable pinhole decreases axial resolution because more out of focus fluorescence contributes to the image. Closing the pinhole more than one Airy disk reduces the light intensity detected because 1 Airy disk contains approximately 80% of the light from a point source detected by the objective lens. dz ( m) dz ( m) a) Numerical Aperture 1.4 b) Pinhole in Object Plane ( m) Figure 40: Influence of the aperture (a) and the pinhole size on the axial resolution. a) The axial resolution is shown in dependence of the NA of the objective lens. b) The axial resolution for an objective with a NA of 1.25 in function of the pinhole diameter projected onto the object plane. The larger the pinhole is the lower the resolution in axial direction because more out of focus fluorescence contributes to the image. 29

30 Fluorescence microscopy Spinning disk microscope Spinning Disk microscopes project the excitation light through hundreds of circular apertures simultaneously into the sample and collect the returning emission through the same confocal apertures. The beams are not actually scanned but moved by spinning the disk rapidly to illuminate the whole field of view. Spinning Disk systems are a form of parallel confocal microscope, meaning they work on multiple areas of the sample at the same time, in this way the main gain of these microscopes is speed. Figure 41. Scheme of a spinning disk microscope. Lasers Multiple lasers can be combined in a modern CLSM and used to excite various fluorochromes. The laser power is adjusted either by neutral density filters or by acousto-optical tunable filters which are driven electronically to individually transmit only a fraction of certain wavelengths. The laser beam is expanded to fill the back aperture of the objective and forms an intense diffraction limited spot that is scanned from side to side and from top to bottom over the specimen in a pattern called a raster. This procedure is called point scanning. Photomultiplier tube detector Fluctuations in light intensity are converted into a continuously changing voltage by the photomultiplier tube detector (PMT). The analogue signal is digitized at regular time intervals by an analogue to digital converter to generate pixels (digital picture elements) that are stored in the computer and displayed on the monitor. 30

31 Fluorescence microscopy Figure 42: a) beam path, lasers and spectral detectors in a modern CLSM (Leica SP2). Please not the schematic drawing of a accousto optical beamsplitter on the right (circle) and the selection of a band of wavelenths after the prism by a variable slit (circle left). Advantages of CLSM The principal gain in imaging using CLSM is the ability to optically section through fluorescent objects (up to 50 µm thick) and to obtain at the same time high contrast images. a) b) Figure 43: Comparison of a widefield (a) versus CLSM (b) image. Images were acquired in three dimensions and projected to one plane. Please note the higher signal to noise and lower background in the CLSM image. 31

32 Multiphoton microscopy MULTIPHOTON MICROSCOPY A fluorescent molecule can be induced to fluoresce by the absorption of a single photon of proper excitation wavelength. The same energy can be provided by two (or more) photons having half (or a fraction of the number of absorbed photons) the energy of a single excitatory photon. If these wavelengths (energy of a photon corresponds to the frequency) are absorbed quasi simultaneously by the fluorochrome, the electron is excited to a higher level (Figure 44). Figure 44: Principle of two photon excitation Multiphoton microscopy has several advantages compared to classical one photon laser scanning microscopy: Excitation is in the near infrared. Infrared radiation is less scattered and absorbed by tissue compared to visible light and deeper penetration depth are possible. This allows in vivo imaging up to 1 mm in brain compared to approximately 100µm in CLSM imaging (Figure 45). Fluorochromes in the focal plane are differentially excited because with multiphoton excitation the efficiency of fluorescence excitation depends on the square of the light intensity, not directly on the intensity as is the case for one photon excitation in CLSM. Thus, only fluorochromes in the focal plane are excited and not in the whole thickness of the specimen (Figure 46). Figure 46. One versus two photon excitation of fluorescein. Please note the very small excitation volume only in the focal plane in two photon excitation. 32

33 Multiphoton microscopy This is called the two-photon advantage and a hallmark of multiphoton microscopy. No confocal detector pinholes are needed, because the fluorescence is only generated in the focal plane (see Figure 65). Additionally all emission light can be detected even light that has been scattered and, therefore, seems to come from different locations (Figure 47). Multiphoton microscopy to occur needs that photons meet "simultaneously" in space and time. For the space, high spatial densities of photon can be achieved by focusing the beam through a high numerical aperture objective. For the time, special lasers are needed to achieve multiphoton absorption. The probability to absorb two photons is less than 1 : compared to absorb one photon. To this end, pulsed infrared lasers are employed which deliver laser pulses of 100fs with 100 GW (!) peak power. The average power is approximately 3 W with a pulse repetition rate of 90 Mhz. Figure 47. Signal collection in clear tissue (no scatter) and in scattering tissue (scatter). Disadvantages of multiphoton microscopy are the high costs of lasers (> CHF), the relative complicated setup with vibration free optical table and the still relative complicated operation of lasers. Effect of light on living cells Energy is related to wavelength. E = hc/λ Therefore, shorter wavelengths are more energetic than long ones. Especially UV (200nm 400nm) illumination of cells is damaging, because photons are energetic enough to break covalent bonds, thereby creating reactive free radicals. IR light (700nm 1000nm) has less energy but is strongly absorbed by carbon bonds leading to accumulation of heat. Visible light is absorbed poorly by cells especially in green and yellow. For extended imaging of living cells light in the green spectrum is beneficial. It is also apparent that cells must be protected from unwanted UV and IR radiation by special filters. These filters can be placed before the light source. Phototoxicity can be recognized by the cessation of cell motility, arrest of organelle movement and cytokinesis. If exposure to light is especially high cells respond by blebbing and swelling. Excessive heat build up can be omitted by imaging at lowest light possible. Additional chemical protection may be necessary against reactive free radicals. 10 mm ascorbate or succinate in the cell culture medium can neutralize free radicals. It is also know that oxygen scavenging enzymes, such as catalase, glucose oxidase can slow down radical build up. 33

34 Detection systems DETECTION SYSTEMS In widefield microscopy and spinning disk confocal systems (disk with many excitation and emission pinholes) CCD cameras are mainly used for detection and acquisition of images. The advantage of using CCD cameras in imaging is the low noise, high sensitivity and resolution of these cameras. Especially when cooling the CCD chip to low temperature low noise and long exposure times are possible. This is useful for fluorescence imaging. By using electron multiplying technology or intensifier tubes CCD camera have an unsurpassed sensitivity making them the primary choice for live cell experiments. In Figure 47 a scheme is shown of how a CCD in principle works. Figure 48: CCD array. A matrix of light sensitive pixel element is illuminated which results in the generation of electrons. After illumination the electrons are transferred to the storage array part of the chip. From this area the electrons are transferred line by line to the serial register from where they are read out pixel by pixel. In point scanning confocal laser scanning systems photomultiplier tubes (PMT) are used to detect fluorescent light. The fluorescent light is gathered from one illuminated point of the specimen at one time. An image is generated by assembling the measured light intensities according to the x, y coordinates which was illuminated. A scheme of a PMT is shown in Figure 49. Figure 49: Photomultiplier tube. Light from a point source in a confocal laser scanning microscope generates electrons which are multiplied by dynodes. This multiplication is of importance due to the low intensity of fluorescent light generated by the small volume excited and the small time the spot is illuminated in a point scanning confocal laser scanning microscope. 34

35 Special fluorescence microscopy applications SPECIAL FLUORESCENCE MICROSCOPY APPLICATIONS Fluorescence after photobleaching Fluorescence recovery after photobleaching (FRAP), a relatively old and simple technique, is used to measure the mobility of molecules. A defined area of fluorescently labelled molecules is bleached with intense illumination typically using laser light. Subsequently, the recovery of fluorescence is recorded over time. The measured recovery of fluorescence can be fitted with appropriate models (diffusion, transport) leading to diffusion constants, rate of transport or fractions of immobilized molecules. Figure 50: Fluorescence recovery after photobleaching of β-adrenergic receptors labelled with Alexa 488 in the cell membrane of COS-7 cells. An intense pulse of light was used to bleach an area (blue square) and the reappearance of fluorescent molecules inside that region was monitored (red square) reflection diffusion and transport. The intensity over time is plotted and analyzed (graph on the left) Fluorescence resonance energy transfer Fluorescence resonance energy transfer (FRET) microscopy is a technique to probe the distance of two fluorescently labelled molecules. FRET occurs if the excitation spectrum of the acceptor molecule overlaps the emission spectrum of the donor molecule and if both, acceptor and donor, are in close proximity. The distance for FRET to occur is stringently limited to the so called Förster distance of the two molecules. Förster distances are in the range of up to 10 nm therefore resulting in much preciser indication interaction than colocalization of two differently labelled molecular species. Examples of suitable donor / acceptor pairs are CFP / YFP for genetically encoded fluorescent proteins or Alexa 488 and Alex 594. Figure 51. Principle of FRET. In the left image, donor (blue) and acceptor (yellow) are not close enough for resonance energy transfer to occur. In the right image, the donor transfers its energy radiation free to the acceptor and FRET occurs. The distance of the donor acceptor in this case is in the range of the Förster distance. 35

36 Deconvolution DECONVOLUTION Deconvolution is a mathematical operation to recover an image that is degraded by blurring and noise. In fluorescence microscopy the blurring is largely due to diffraction limited imaging by the instrument. All images are subject to some sort of degrading process. Consider as a 2D example a moving camera that creates a vague picture (the measured image) of a scene (the object). Here the camera displacement is an a priori known degrading process. Restoration applies the inverse process on the degraded image in order to recover the true object. In microscopy we apply the same technique on 3D images where the degrading process is the diffraction and aberration of the microscope lens. The image formation process is mathematically described by a convolution equation of the form (Eq.1) where the image g arises from the convolution of the real light sources f (the object) and the Point Spread Function (PSF) h. In other words, the microscope yields an image g, which is a degraded version of the object "f". The degradation is caused by blurring (convolution) by the PSF, wider PSFs leading to more severe blurring. The PSF can be measured, for example by recording beads, or by a theoretical optical calculation based on knowledge of the microscopic parameters. This is outlined in the following illustration: Figure 52. The aim of deconvolution is to recover the original object "f"

37 Super Resolution Microscopy SUPER RESOLUTION MICROSCOPY For centuries, cell biology has been based on light microscopy and at the same time been limited by its optical resolution. Thus, optical microscopes cannot visualize details much finer than about half the wavelength of light. However, several new technologies have been developed that bypass this limit (Schermelleh et al., 2010). Structured illumination microscopy SIM. Figure 53. The sample plane is excited by a nonuniform wide-field illumination. Laser light passes through an optical grating, which generates a stripe-shaped sinusoidal interference pattern. This combines with the sample information originating from structures below the diffraction limit to generate moiré fringes. The image detected by the CCD camera thus contains high spatial frequency sample information shifted to a lower spatial frequency band that is transmitted thought the objective. A mathematical reconstruction allows, from a series of raw images per slice to reconstruct a high-resolution image with doubled resolution in xy compared with widefield resolution. In 3D-SIM additional doubling in the axial resolution is achieved. Stimulated emission depletion microscopy STED. Figure 54. The focal plane is scanned with two overlapping laser beams. While the first laser excites the fluorophores, the second longer wavelength laser drives the fluorophores back to the ground state by the process of stimulated emission. A phase plate in the light path of the depletion laser generates a donut-shaped energy distribution, leaving only a small volume from which light can be emitted that is then being detected. Thus, the point spread function (PSF) is shaped to a volume smaller than the diffraction limit

38 Super Resolution Microscopy Single molecule localization microscopy PALM-STORM. Figure 55. These microscopy techniques assure that only a relatively low number of fluorophores are in the emitting (active) state. This is achieved either by photoactivation, photoswitching, triple state shelving, or blinking. These molecules are detected on the CCD camera as diffraction-limited spots, whose lateral position is determined with very high accuracy by a fit. Single molecule positions from several thousand raw images, each with a different subset of emitters, are then used to generate a density map featuring several hundred thousand single molecule positions within the plane of focus. Literature Fundamentals of light microscopy and electronic imaging, Douglas B. Murphy, Wiley-Liss Light and Electron Microscopy, Elizabeth M. Slayter, H S. Slayter, Cambridge University Press Fine Structure Immunocytochemistry, G. Griffith, Springer-Verlag. Handbook of biological confocal microscopy, James B. Pawley, Plenum Press. Molecular biology of the cell, Alberts et al., 4th Ed. Garland Science, Schermelleh L., Heintzmann R., leonhardt H., A guide to super-resolution fluorescence microscopy (2010) JCB 190: