Confocal fluorescence microscopy has been widely used for observing three-dimensional (3D) high-resolution fluorescent images in cells or tissues [1,2]. As a method of visualizing fluorescent images in cells or tissues using a single photon, confocal microscopy can penetrate up to 100 μm within the tissue. In contrast, the more recently developed two-photon confocal imaging method focuses two near-infrared photon beams with longer wavelengths of around 700 to 1,000 nm, enabling the imaging of tissues of up to 1,000 μm in depth [3].

Since two-photon confocal imaging uses photons of long wavelengths and thus low energy to obtain image data, it causes less phototoxicity than conventional methods; therefore, imaging of both deeper depths and longer periods is possible [4]. Compared to traditional histological methods, which involve the isolation of tissues or cells from an organism followed by chemical treatment and ultimately imaging in a set period of time, intravital imaging using two-photon microscopy enables the researcher to study not only the structure but also the function of the target cell or tissue of a live animal. Taking advantage of such properties, much spotlight has been given to long-time 3D imaging of organs or tissues in live animals, mainly mice. To visualize target cells or tissue, mice are genetically modified to express fluorescent proteins or stained in specific cells and molecules or fluorescent antibody is used to label the target molecule. Intravital imaging using two-photon microscopy on animal models is an extremely useful technique that allows for real-time observation of cell movement and structure within the tissue and is widely utilized in immunology. It also enables visualization of cell movement and structure on a single-cell level, thus making direct observation of immune cell intravascular migration, extravasation, and interstitial migration possible, along with intracellular molecules that are involved in cell movement during such processes.

Thus, unlike conventional single-photon imaging methods, intravital imaging using two-photon confocal microscopy reveals more accurate and detailed information on cell-level movements within organisms and provides knowledge on the physiological and pathological roles of immune cells in inflammations and other diseases. As such, this imaging tool is being widely utilized in basic research that pursues novel therapeutic solutions.

For more than a decade, pioneering research in fields such as immunology has utilized intravital imaging to yield significant results. Especially, due to the ability of two-photon microscopy to penetrate up to 1,000 μm in depth, intravital imaging has been frequently used in virtually every organ in research on mouse models of disease—including the respiratory system, lymphatic system, cremaster muscle, liver, and linings of the stomach, pancreas, kidney, spleen, heart, brain, and dermal tissue [5-9]. Depending on the study purpose, consecutive intravital imaging within various time periods is possible, ranging from 20 minutes to several hours.

On the other hand, one organ of interest can also be repeatedly imaged in the span of a few days. Therefore, long-term intravital imaging enables the observation of pathological and physiological phenomena involved with specific organs or tissues. Despite such advantages, two-photon confocal microscopy requires expensive equipment. In addition, even after acquiring such machinery, the actual surgery that exposes the organ of interest, along with methods of fixing the anesthetized organism to minimize the effect of heart pulsation and respiration, is also a task for the researcher. As of now, animal fixation equipment for brain tissue intravital imaging can be purchased; intravital imaging of other organs and tissues is mainly carried out with self-devised surgical methods and fixation equipment.

To study the biological phenomena that accompany inflammation, researchers have devised several different animal models. In particular, three infection methods have been frequently utilized to induce inflammation in mice; these include viral and bacterial infection as well as lipopolysaccharide (LPS) inoculation. For influenza viral infection, anesthetized mice are intranasally inoculated with an influenza A virus suspension. Intravital imaging of the trachea or the lungs can be performed between 1 and 5 days after infection, depending on the prognosis and immune cell subsets subject to investigation [10].

Similar to influenza viral infection, bacterial infection of the airway through the respiratory tract can be fatal. There are two common nosocomial pathogens that cause respiratory infection: Pseudomonas aeruginosa and Staphylococcus aureus. P. aeruginosa is an encapsulated, rod-shaped, Gram-negative bacterium that possesses the ability to cause opportunistic lung infections in patients with impaired immunity. It is wellknown that these infections lead to high mortality and morbidity in cystic fibrosis patients [11]. Therefore, these bacterial infection models of the airway of live mice would be of help to investigate actual responses of immune cells through intravital imaging. In general, wild-type P. aeruginosa strains, PAO1, and human-derived PA14 are generally used in animal infection models. The PA14 strain has a wide spectrum of infections from humans to animals and is virulent, unlike PAO1 [12]. The other common source of respiratory infection, S. aureus, is a round-shaped, Gram-positive bacteria that has several virulence factors including toxins and enzymes. Lipoteichoic acid and staphylococcal peptidoglycan are major components of S. aureus and are recognized by immune cells in a TLR2-dependent manner [13]. P. aeruginosa and S. aureus infection are correlated with the innate immune system in murine models; pathogen recognition and bacterial clearance of innate immune cells may play an important role in controlling P. aeruginosa and S. aureus [14]. It is very useful to perform intravital lung imaging of live animals for the investigation of the direct interaction between pulmonary innate immune cells and P. aeruginosa or S. aureus. From two-photon intravital microscopy, real-time dynamics of innate immune responsive phenomena for host defense such as neutrophil extracellular traps can be measured, which provides insight regarding the response mechanisms to infections [15,16].

Furthermore, LPS has also been frequently employed as a perturbation element to the immune system. LPS, a major component of the outer membrane of Gram-negative bacteria, generates a response that accompanies cytokine activation from the immune system, contributing to the induction of sepsis [17,18]. Thus, the LPS-mediated mouse inflammation model can be used to confirm the activation of circulating leukocytes, migration of immune cells to tissues, and stimulation of these cells for cytokine release [19].

In mice lung inflammatory models, the two main methods of LPS injection are intratracheal and intranasal injection, respectively. While both methods are utilized to deliver the vehicle into the lungs, there exist differences in the injection procedure. Intratracheal injection is an administration method that allows a small amount of the vehicle to be directly delivered to the lungs or even to a single lobe in the lungs via bronchial catheter insertion. This method is particularly effective because of direct administration to the lungs, avoiding deposition of the vehicle in the nasal cavity [20]. On the other hand, intranasal injection takes advantage of respiration, using mice inhalation for vehicle delivery. This method is performed by gradual injection of the vehicle into the nostril of the mouse using a micropipette after anesthesia.

It is a crucial step in intravital imaging to exteriorize the target organ from the live animal. Not only should the proper organ be selected but also surgical methods that cause minimal harm to the organ deserve much scrutiny and practice. In terms of visualizing the respiratory system of mice, researchers often target either the trachea or lungs for imaging, depending on the specific goals of each study [10,21]. For intravital imaging of the mouse trachea, the anesthetized mouse is subsequently placed on a custom-designed platform for imaging. The trachea is exteriorized and a small cut is made on the frontal wall to insert an 18-gauge blunt-end cannula (Figure 1A) [10]. Various mouse strains can be used for exclusive visualization of the motility of particular immune cells; on the other hand, immune cell-specific fluorescent antibodies could also be intravenously injected to monitor circulating immune cells. Such cell visualization and trafficking methods will be discussed in more detail later.

Lung exposure is performed via thoracotomy, in which the thoracic cavity is opened to image the lung tissue. It is important to note that, while the thoracic cavity is opened, the pressure in the lung changes, making spontaneous breathing of the mouse difficult. Thus, prior to thoracotomy, intubation using an intravenous catheter is required [21,22]. A continuous supply of oxygen and isoflurane flows through the inserted catheter for life support during imaging. After intubation, skin and fatty tissue are removed to expose the thorax, the ribs are opened with surgical forceps, and the imaging apparatus is inserted. Finally, light vacuum pressure is applied to the apparatus to adsorb the lungs to the imaging window, and intravital imaging is conducted (Figure 1B).

To observe the movement and morphology of immune cells using intravital imaging, the labeling of immune cells is necessary. Numerous mouse strains have been generated and utilized for research, depending on the cell subset to be imaged. Lysozyme M-green fluorescent protein (LysM-GFP) mice have GFP attached to a particular type of lysozyme, which is an immune enzyme that degrades peptidoglycans in bacterial cell walls (Figure 2A). Lysozyme M, encoded by the Lyz2 gene, is expressed predominantly in neutrophils and also partly in macrophages; therefore, this strain is generally used for neutrophil visualization [23,24]. Another transgenic mouse line, the CX3Cr1-GFP mouse, is a strain with GFP attached to the CX3C chemokine receptor 1 (CX3Cr1), which is utilized as a marker for macrophages and monocytes [25,26]. Thus, CX3Cr1-GFP mice are used for fluorescent visualization of monocytes and macrophages. While fluorescent labeling can be accomplished with the aforementioned methods, in some cases, it is prudent to visualize the flow of blood itself. Dextran is a commonly used blood vessel labeling material; for example, in mouse strains with GFP-labeled immune cell populations, Texas Red dextran could be used to stain blood vessels to create images with sufficient contrast between labeled immune cells and blood flow [27].

To separately visualize blood vessels and leukocyte subtypes as described above, intravenous injection of fluorophore-conjugated dextran is widely used. Among various dextrans of different molecular weights, 70 kDa dextran, which has a molecular weight similar to that of bovine serum albumin (66.5 kDa), is particularly useful in visualizing blood vessels due to its tendency to flow along with blood circulation; in terms of color, injection of fluorophore (Texas Red [Ex_max, 596; Em_max = 615], fluorescein isothiocyanate [FITC; Ex_max, 490; Em_max, 525])—conjugated dextran is specifically utilized for bloodstream visualization. While such tools offer a powerful method to image blood vessels and flow, vascular leakage from blood vessels to interstitial tissue could occur at stimulated or damaged vessels during the course of intravital imaging using fluorophore-conjugated dextran. Taking advantage of this phenomenon, the intensity of vascular leakage could also be quantified with the proper analysis software (e.g., Volocity, Imaris, ImageJ) [28].

Antibody or Labeling Dye to Leukocyte Subtypes Another method of visualizing leukocyte subsets besides genetic strains is to use fluorescent antibodies or labeling dyes. Fluorescent antibody injection makes use of the property of antibodies to bind to specific antigens (Figure 2B, C). An example of a cell-surface protein that is abundantly expressed in murine neutrophils is Ly6G [29], while those expressed in macrophages or monocytes include Ly6C and F4/80 [30]. On antibodies that bind to these antigens, fluorescent dyes such as Alexa Fluor 488, FITC, or phycoerythrin are conjugated, allowing for separate visualization of cells according to wavelength [31].