Chapter 20 Experimental Systems and Methods

Immunofluorescence-Based Imaging Techniques

The last decade has seen a veritable explosion in the manner in which fluorescence microscopy has been applied to studies of the immune system. Since students will be exposed to many of these techniques in the current primary literature, we present here a summary of the common variations of immunofluorescence. See also Advances Box 20-1, which introduces students to dynamic imaging techniques that visualize cells in real time during an ongoing immune response and offers hints on how to view the resulting movies with an educated and critical eye.

ADVANCES BOX 20-1

Dynamic Imaging Techniques, or How to Watch a Movie

As described in this chapter, dynamic imaging, otherwise known as intravital microscopy, allows investigators to image living tissues such as the popliteal or inguinal lymph node, or a section of the intestine of a living, anesthetized mouse. Videos of cell trafficking in intact tissue are now relatively common additions to published papers. However, videos, no less than data that appear in figures and tabular form, must be viewed with a critical eye. We suggest that students will benefit considerably if they ask the following questions, when viewing movies provided as part of immunological papers:

Which cells are labeled?
For technical reasons, investigators can typically label only two or three cell types at a time, and this means you are seeing only a fraction of the cellular interactions that are occurring in the imaged tissues. Apparently empty areas in a video may, in reality, be dense with unlabeled cells and extracellular material. For example, it was first thought that T and B cells moved through lymph nodes simply by following soluble gradients of chemokines. However, both static and dynamic imaging data have now revealed networks of reticular cells and fibers on which the cells travel in ways that are assisted by chemokine gradients. Therefore, always ask yourself what isn’t there; not just what is.
What fraction of the population of interest carries the label?
In some experiments, labeling all the cells of interest would lead simply to a blur of label, so investigators sometimes elect to label only a small fraction of their input population. Although the activity of this fraction is most probably representative of all similar cells, the picture we see in our image should be modified in our heads to reflect reality. In addition, any subsequent calculations regarding the ratios of the relative numbers of cooperating cells (e.g., T cells, or dendritic cells, reacting with B cells) will have to take into account the fraction of cells that carry the original label.
How are cells labeled and how stable is the label?
If cells are labeled with a fluorescent protein such as GFP or YFP, make sure you understand the properties of the promoter/enhancer system that drives its activity. In Figure 1a, the promoter that is driving YFP drives expression of the CD11c antigen, so any cell that bears CD11c will also glow yellow, even if the figure legend tells you that only DCs are labeled. Alternatively, if the cells are labeled with fluorescent antibodies before infusion, information should be obtained about the specificity and cross-reactivity of the antibody that is used in the labeling procedure, so you can understand exactly what cells and tissues will be labeled. Another variable to consider is the stability of the label. The apparent loss of fluorescent cells could mean the fading of a stain, particularly if illumination is prolonged and the dye is subject to quenching. Alternatively, the loss of fluorescence could result from the reduction in expression of a fluorescent protein after cell proliferation or proteasomal destruction, and not necessarily the loss of a whole cell population. Figure 1b shows an experiment in which T cells were labeled with CFSE prior to injection, allowing investigators to track not only the location of the injected T cells, but also the frequency of their cell division.

FIGURE 1 Fluorescence labeling of immune cells. (a) In this transgenic mouse, the gene for YFP is downstream of the CD11c promoter. Two-photon intravital microscopy was used to visualize the movements of dendritic cells (yellow) in the skin after infection with Leishmania major. The extracellular matrix can be seen because it produces what are known as “second harmonic” fluorescence signals under the illumination of the incident light. The migration of these cells can be visualized in this YouTube video: www.youtube.com/watch?v=XOeRJPIMpSs. (b) Cells (e.g., T cells) can be labeled with membrane-permeable dyes that fluoresce green (CFSE) or red (CMTMR) and injected intravenously (adoptively transferred). Miller and colleagues were among the first to visualize CFSE-labeled T-cell behavior in a lymph node, a static example of which is shown here with HEV labeled red.
When and how are cells introduced?
Most current dynamic imaging systems involve the introduction of fluorescently tagged cell populations into a recipient mouse. However, manipulations of cells performed in vitro, prior to introduction, can alter cell function and subsequent cell migration. Antigen-presenting cells, for instance, can be nonspecifically activated in culture, which can change their trafficking patterns and cell interactions. Awareness of the potential artifacts introduced by in vitro manipulations will improve your ability to understand and critique the videos. In addition, you should always look for, and understand the rationale behind, the timing and sequence with which cells are introduced into an animal.
When and how is antigen introduced?
The ability to track cells during an immune response requires that antigen be introduced. Antigen can be noninfectious—for example, a foreign protein such as ovalbumin, for which transgenic T and B cells are specific—or infectious (e.g., Toxoplasma gondii; lymphocytic choriomeningitis virus, LCMV). Antigen introduction by infection with living pathogens will inspire a natural, full-fledged innate immune response that can in turn alter cell migration patterns. Antigens can also be introduced by a variety of routes, and the mode of introduction can dramatically influence the quality and outcome of the immune response. For example, respiratory system immunity is best generated by introducing antigen intranasally. Antigen may also be given in the presence or absence of adjuvant or cytokines. Be particularly careful to take the nature and mode of introduction of antigen into consideration when these are not the primary topics of interest of the investigators, as it is then that potential artifacts are most likely to occur.
How are investigators visualizing antigen-specific immune cells?
In order to understand the behavior of antigen-specific lymphocytes, one has to be able to distinguish them from the millions of other lymphocytes that are not specific for the particular antigen of interest. Investigators will often introduce fluorescent lymphocytes whose antigen specificity is known (e.g., cells isolated from mice that express transgenic T-cell receptors or B-cell receptors). This approach allows one to specifically trace the behavior of antigen-specific cells. However, transgenic systems are not available for every antigen of interest and can introduce their own artifacts (e.g., receptor levels tend to be high, and high cell frequencies do not mimic physiological responses). Investigators have found some clever ways around this by introducing into the genome of the pathogen an antigen against which receptor transgenics react (e.g., ovalbumin or ovalbumin peptide.) They have also found ways to infer antigen specificity from behavior after introducing T cells that are generally fluorescent.
What type of tissue is being examined, and under what conditions?
Lymph nodes are easier to image by two-photon intravital microscopy than is the spleen, which has more structures that autofluoresce (i.e., generate a fluorescence signal in the absence of addition of a label). The temperature, humidity, and culture conditions for the lymph node will have a significant influence on cell trafficking, and therefore isolated organs must be maintained in perfusion chambers that stabilize these conditions. However, even with these variables accounted for, cellular traffic patterns in explanted organs may not reflect all that is physiological since the surgery required to image the node can itself inspire inflammation and stress that in turn influence trafficking.

Questions Dynamic Imaging Allows Us to Ask and Answer

Dynamic imaging is now a vital component of an immunologist’s experimental arsenal. It can offer visually stunning confirmation or refutation of predictions suggested by more indirect and static approaches, and it can also provide information about the sequencing of immune cell interactions within a microanatomical context that could not otherwise be gained. Dynamic imaging has already revealed unexpectedly important sites of immune-cell interaction (e.g., the subcapsular sinus of a lymph node), unexpected participants (neutrophils) in lymph node activity, and unique shape changes and motility patterns (e.g., within germinal centers) that challenge or enhance conventional immunological wisdom. It has allowed investigators to directly address centrally important questions: Which cells do naïve T cells encounter first during an immune response? What types of interactions are associated with activation, death, or tolerance? And ever more specific questions can be asked. For instance, do somatically mutating dark-zone germinal-center B cells travel often to the light zone to test their antigen specificity? Furthermore, dynamic imaging is also moving beyond the lymph node into studies of barrier immunity and intestinal vasculature in a way that opens up new opportunities to investigate intestinal pathogens.

However, even these striking imaging experiments suffer from technical limitations. Observations can be made only in one or two organs of a mouse in a single experiment, and only for 20 to 60 minutes at a time. Dynamic imaging is dependent on experimental systems that have been extensively manipulated, and the antigen-specific cell frequencies observed in such experiments are usually significantly higher than they are under physiological situations.

Nonetheless, dynamic imaging cannot be overestimated as an educational tool that inspires as effectively as it teaches. The immune system is a marvelous tangle of multidimensional complexity and organization, and a simple video can clarify in minutes what our rich yet imperfect language cannot easily articulate. Richard Avedon, a twentieth-century American photographer, perhaps said it best: “All photos are accurate. None of them is the truth.” However, moving images of cells in multidimensional time and space may inspire us to move ever closer to the truths inherent in our discipline.

Fluorescence Can Be Used to Visualize Cells and Molecules

The phenomenon of fluorescence results from the property of some molecules (fluorescent dyes) to absorb light at one wavelength and emit it at a longer wavelength. If the emitted light has a wavelength in the visible region of the spectrum, the fluorescent dye can be used to detect any molecules bound by that dye.

Some fluorescence imaging experiments take advantage of the fact that particular dyes specifically bind to particular macromolecules. For example, the blue dye 4′,6-diamidino-2-phenylindole (DAPI) specifically stains DNA. Other protocols utilize the affinities of easily obtained proteins (which can be readily conjugated with fluorescent dyes) to bind to biologically important molecules. For example, the protein phalloidin, which specifically binds filamentous actin, can easily be conjugated to fluorescent probes. Similarly, the soluble protein annexin A5 binds to phosphatidylserine, which is exposed on the outer surface of cells undergoing apoptosis; fluorescently labeled annexin A5 is easy to obtain and use as a measure of apoptosis. In immunofluorescence measurements, antibodies or streptavidin can be artificially conjugated to a host of dyes. These various approaches to fluorescence labeling can be combined with one another to provide spectacular images of cellular and subcellular structures (Figure 20-14a).

A two-part figure shows fluorescence micrograph of human dendritic cell with fluorescent-labeled Listeria monocytogenes in section a. Section b shows schematic diagram of fluorescence microscope. — FIGURE 20-14 Fluorescently labeled cells and the passage of light through a fluorescence microscope. (a) Fluorescence microscopic image of a human dendritic cell infected with engineered *Listeria monocytogenes*. The green fluorescence is generated by Alexa Fluor 488–conjugated phalloidin (which binds actin filaments). Red fluorescence derives from red fluorescent protein expressed by the bacterium. DNA is stained with DAPI (blue). (b) Light from a source passes through a barrier filter that only allows passage of blue light of particular wavelengths. The light is then directed onto the sample by a dichroic mirror that reflects light of short wavelengths (below approximately 510 nm) but allows passage of higher wavelengths. When the blue light interacts with the sample, any fluorescent molecules excited by it emit fluorescence that then passes through the dichroic mirror, through a second barrier filter, and then is transmitted to the eyepiece.

The schematic of fluorescence microscope has

First barrier filter that lets through only blue light with a wavelength between 450 and 490 nanometer

Blue light passing through the first barrier filter hits the beam-splitting mirror that reflects light below 510 nanometers, but transmits light above 510 nanometers. This mirror is placed at 45 degree angle to the objective lens at the bottom of the fluorescence microscope.

Light reflected from the beam-splitting mirror passes to the objective lens, below which the object is placed

Light from the object passes through the objective lens, through the beam-splitting mirror, to the Second barrier filter placed just below the eye piece. The Second barrier filter cuts out unwanted fluorescent signals passing the specific green fluorescein emission between 520 and 560 nanometers.

Such images can be visualized by fluorescence microscopy, which uses short-wavelength light to excite the fluorescent dyes. A series of filters and mirrors that can be adjusted by the investigator can then be employed to select which wavelengths of light will reach the eyepiece and therefore which molecules will be visualized (see Figure 20-14b). Modern instruments use combinations of several filters and mirrors that allow the investigator to detect light emitted at multiple different fluorescence wavelengths. The different colored images generated from the various dyes can then be overlaid by the instrument’s software to provide a single representation in which the locations of the antibody-bound molecules can be compared.

In addition to artificially synthesized dyes, some naturally occurring proteins such as GFP and RFP (green and red fluorescent protein, respectively) contain fluorescent chromophores. Figure 20-15 shows a striking image of three neonatal mice engineered to express GFP under the control of an actin promoter, along with their non-GFP littermates. In some experiments, investigators will place the fluorescent protein genes under the control of particular promoters or make fusion proteins that contain both the native protein and GFP sequences. These adaptations allow researchers to use fluorescence imaging to determine where and when proteins under the control of those same promoters are expressed.

A photo shows 6 neonatal transgenic mice huddled together. Three mice have normal skin while the remaining three mice have fluorescent skin. — FIGURE 20-15 Fluorescence labeling of whole animals with green fluorescent protein. Three neonatal transgenic mice express GFP under the control of an actin promoter.

Confocal Fluorescence Microscopy Provides Three-Dimensional Images of Extraordinary Clarity

One of the limiting factors in obtaining clear images by fluorescence microscopy is that fluorescent molecules lying above and below the focal plane can contribute to the light that reaches the objective, leading to a blurred image. In confocal microscopy, that artifact is eliminated by using an objective lens that focuses the light from the desired focal plane directly onto a pinhole aperture in front of the detector (Figure 20-16). Light emitted from molecules located at other levels within the sample is stopped at the perimeter of the pinhole, and remarkably clear images of a single plane within the sample can thus be generated. In laser scanning confocal microscopy, investigators use lasers to provide the exciting light and computing power to move the focal plane in all three dimensions, thus enabling them to scan an x, y plane at different depths of focus, and reconstitute powerful three-dimensional images.

Schematic diagram of confocal microscope shows the pathways of different focal planes. — FIGURE 20-16 The principle of confocal microscopy. The sample is illuminated by a laser beam that excites fluorescence from dyes in several different focal planes, represented here by the green, red, and purple lines. However, passage of light through the confocal pinhole (shown on the left-hand side of the image) filters out light emitted from all but a single focal plane, resulting in an extraordinarily clear image. Computer control of the exact plane from which light can be received through the pinhole allows the development of images from a number of focal planes and the creation of a composite three-dimensional representation.

The schematic shows the laser beam passing through the point wise illumination through the beam splitter to the objective lens at the bottom of the microscope. On the left side of the image, the confocal pinhole is present. The detector is present on the left of the confocal pinhole. The corresponding text reads, Point wise detection. The sample (x, y, z) is placed below the objective lens. The path of the beam is as follows:

When sample is on the focal plane (in focus) – The laser beam passes through the point-wise illumination through the beam splitter to the objective lens. The reflected beam from the focal plane passes back through the objective lens, reflects from the beam splitter, passes through the confocal pinhole, and falls on the detector.

When sample is placed below focal plane (out of focus) – The laser beam passes through the point-wise illumination, through the beam splitter, to the objective lens. The reflected beam from below the focal plane passes back through the objective lens, reflects from on the beam splitter, and falls on the confocal pinhole.

When sample is placed above the focal plane (out of focus). The laser beam passes through the point-wise illumination, through the beam splitter, to the objective lens. The reflected beam from above the focal plane passes through the objective lens, reflects from the beam splitter, and falls on the confocal pinhole. A double headed arrow labeled Z marks the three sample placement locations.

Multiphoton Fluorescence Microscopy Is a Variation of Confocal Microscopy

Two-photon and multiphoton microscopy are variations on confocal microscopy that offer even greater resolution in the development of three-dimensional images. In standard confocal microscopy, excitation of the fluorescent probes (dyes) occurs along the whole path of the laser beam through the tissue (Figure 20-17a). This means that, although the emission beams are derived only from a single level within the sample, fluorescent probes are being excited throughout many levels of the tissue. Since fluorescent probes will eventually photobleach (i.e., cease to emit light) after extensive excitation, this limits the useful life of the sample. It also means that some additional light is emitted from the sample that must be filtered out of the final image.

Two photos labeled a) 1-photon and b) 2-photon are shown. Both photos show the close view of the objective lens and the fluorescence generated. — FIGURE 20-17 Fluorescence excitation by one-photon versus two-photon laser excitation. (a) On the left-hand side, we visualize excitation by a short-wavelength, high-energy laser that excites fluorescence from all molecules in its path. (b) On the right-hand side, we see the fluorescence that results when molecules are excited with near simultaneity by two beams of longer wavelength, lower energy laser light. Since excitation can occur only in the region where the two beams intersect, it is limited to a single focal plane.

In multiphoton fluorescence microscopy, long-wavelength lasers are used that emit in the infrared region of the spectrum. The laser beams are relatively low energy, and so more than one photon must impinge on a fluorescent molecule in order to provide sufficient energy to excite the electrons. The low energy of these infrared lasers minimizes the extent of photobleaching and enhances the useful lifetime of the sample, since those parts of the sample that interact with only a single laser beam are not typically damaged. Furthermore, the fact that at least two beams of light are required to bring about fluorescence excitation ensures that excitation occurs only within the plane of intersection of the laser beams (see Figure 20-17b). By moving the focal point of excitation within the x and y planes, information about a full optical section can be generated, and that whole process can then be repeated on additional z levels, thus giving rise to a three-dimensional image. In Chapter 14, you have seen some of the powerful images developed with this technique.

Intravital Imaging Allows Observation of Immune Responses in Vivo

Intravital imaging takes advantage of the ability to maintain both the blood and the lymphatic circulations within lymph nodes, sections of intestine, or other organs, after they have been gently lifted out of an anesthetized donor and onto a warmed perfused microscope stage. Using a multiphoton microscope, three-dimensional images of fluorescently labeled cells and structures can be generated and information gleaned about the behavior of immune cells and molecules essentially in vivo. Advances Box 20-1 describes this approach in more detail and Chapter 14 describes significant advances that have been made using this, and related technologies.

Visualization and Analysis of DNA Sequences in Intact Chromatin

Three-dimensional fluorescence in situ hybridization (3-D FISH) is a relatively new technique in which fluorescent RNA probes labeled with several different colors are generated that hybridize with known DNA sequences. For example, probes have been developed that hybridize with V_H sequences known to be situated farthest from the D_H regions (distal V_H sequences), or with V_H sequences closest to the D_H region (proximal V_H sequences). Investigators have used combinations of such probes to study the relative arrangements of BCR and TCR genes in cell nuclei at defined stages of lymphoid development. Sophisticated analytical tools then use the distances measured between the individual probes to develop topological maps of the chromosomes. Figure 20-18 is a striking image that allows us to see with our own eyes how the distance between D-J_H distal and D-J_H proximal V_H genes is altered in pro-B cells as the Ig-encoding chromosome undergoes contraction during V-D recombination.

Two micrographs show the results obtained from 3D structured illumination microscopy of genes in in E L 4 T cells and pro-B cells. The micrographs are captured in scale of 5 micrometer. — FIGURE 20-18 Three-dimensional fluorescence in situ hybridization (3-D FISH). 3-D FISH is used to show the distances between proximal (V_H7183—red) and distal (V_HJ558—green) genes in EL4 T cells (left) relative to those in pro-B cells (right). On the left-hand side, we see that, in T cells, the proximal and distal gene families (red and green, respectively) are at a distance where they can be readily discriminated from one another. In contrast, in pro-B cells, where the Ig chromosome is undergoing contraction, the two gene families are closer together and the two colors can barely be resolved, if at all.