We describe a protocol to surgically expose and stabilize the murine submandibular salivary gland for intravital imaging using upright intravital microscopy. This protocol is easily adaptable to other exocrine glands of the head and neck region of mice and other small rodents.
The submandibular salivary gland (SMG) is one of the three major salivary glands, and is of interest for many different fields of biological research, including cell biology, oncology, dentistry, and immunology. The SMG is an exocrine gland comprised of secretory epithelial cells, myofibroblasts, endothelial cells, nerves, and extracellular matrix. Dynamic cellular processes in the rat and mouse SMG have previously been imaged, mostly using inverted multi-photon microscope systems. Here, we describe a straightforward protocol for the surgical preparation and stabilization of the murine SMG in anesthetized mice for in vivo imaging with upright multi-photon microscope systems. We present representative intravital image sets of endogenous and adoptively transferred fluorescent cells, including the labeling of blood vessels or salivary ducts and second harmonic generation to visualize fibrillar collagen. In sum, our protocol allows for surgical preparation of mouse salivary glands in upright microscopy systems, which are commonly used for intravital imaging in the field of immunology.
Saliva is secreted by exocrine glands to lubricate food, protect the mucosal surfaces of the oral tract, and to deliver digestive enzymes as well as antimicrobial substances1,2. In addition to minor salivary glands interspersed in the oral submucosa, there are three bilateral sets of major glands identified as parotid, sublingual, and submandibular, according to their location1,2. Pyramid-shaped epithelial cells, organized into flask-shaped sacs (acini) or demilunes that are surrounded by myoepithelial cells and a basement membrane, secrete the serous and mucous components of saliva1. The narrow luminal space of the acini drains into intercalated ducts, which unite into striated ducts until they finally join into a single excretory duct1. The main excretory duct of the SMG is called Wharton's duct (WD) and opens into the sublingual caruncle3,4. The SMG epithelial compartment therefore represents a highly arborized structure with manifold terminal endpoints, resembling a bundle of grapes1,5,6. The SMG interstitium is composed of blood and lymphatic vessels embedded in connective tissue7 containing parasympathetic nerves8 and extracellular matrix5. Normal human and rodent salivary glands also contain T cells, macrophages, and dendritic cells9, as well as plasma cells that secrete Immunoglobulin A (IgA) into the saliva9,10. Due to its multifaceted functions in health and disease, the SMG is a subject of interest for many fields of biological research, including dentistry4, immunology11, oncology12, physiology8, and cell biology3.
Imaging of dynamic cellular processes and interactions is a powerful tool in biological research13,14. The development of deep tissue imaging and innovations inmicroscopes based on nonlinear optics (NLO), which rely on scattering or absorption of multiple photons by the sample, has allowed to directly examine cellular processes in complex tissues13,15. Absorption of multiple photons involves delivery of the total excitation energy by low energy photons, which confines fluorophore excitation to the focal plane and thus allows deeper tissue penetration with reduced photodamage and noise from out of focus excitation13,15. This principle is employed by two-photon microscopy (2PM) and allows for imaging of fluorescent specimens in depths of up to 1 mm15,16. While commercially available 2PM setups have become user-friendly and reliable, the major challenge for intravital imaging is to carefully expose and stabilize the target organ of anesthetized mice, especially for imaging of time lapse series. Several methods for digital drift correction after data acquisition have been published17,18 and we recently developed "VivoFollow", an automated correction system, which counteracts slow tissue drift in real time using a computerized stage19. However, it is still critical for high quality imaging to minimize tissue motion, especially fast movements caused by breathing or heartbeat19. Preparation and stabilization procedures have been published for multiple organs, including spinal cord20, liver21, skin22, lung23, and lymph node24. Furthermore, models for rat salivary gland imaging have been developed3,25 and further refined for high resolution intravital imaging of the murine SMG tailored to an inverted microscope setup26,27,28.
Here, we present a practical and adaptable protocol for intravital imaging of the murine SMG using upright nonlinear microscopy, which is commonly used for intravital imaging in the field of immunology. To this end, we modified a widely employed immobilization stage used for popliteal lymph node preparations.
All animal work has been approved by the Cantonal Committee for Animal Experimentation and conducted according to federal guidelines.
1. Anesthetize the Mouse
2. Remove the Fur from the Operating Area
3. Fix the Mouse on the Stage
NOTE: This protocol uses a customized stage (Figure 1) equipped with two freely adjustable holders for cover slips (one attached to the stage with a screw, the other fixed permanently onto it); adjustable stereotactic ear bars; a string, which can be tightened with a screw; a freely adjustable metal heating ring; and a raised area designated to place the torso of the mouse. This stage is modified from a popliteal lymph node stage, with an added stereotactic holder and a support for the lower cover slip holding the exteriorized SMG. Detailed plans or quotes of the stage are available upon request.
4. Surgical Exposure of the Salivary Gland Using a Stereomicroscope
NOTE: The following steps require the use of a stereomicroscope, fine forceps, and surgical scissors. Since the procedure is terminal, it is not necessary to maintain sterile operating conditions. Ethanol-cleaned instruments suffice.
5. Fixing the Gland
6. Fixing the Heating Ring and Temperature Probe
7. Imaging
NOTE: We use an upright two-photon microscope system equipped with a tunable Ti:Sa Laser and a fluorescent microscope with water-immersion 20X or 25X objectives. Real time correction of tissue drift during acquisition can be implemented by using an automated stage and VivoFollow19. This improves the stability of image acquisition, but is not required.
This protocol allows imaging of almost the entire dorsal or ventral side of the SMG. The field of view typically also includes the sublingual salivary gland, which differs slightly from the SMG in cellular composition4. Both glands are encapsulated by fibrillar collagen and subdivided into lobes. Most 2PM systems can produce a label-free image of fibrillar collagen by measuring the 2nd harmonic signal, but usually fluorescent molecules are needed for visualization of cellular and subcellular structures. For instance, ubiquitous transgenic expression of green fluorescent protein (GFP) in combination with the 2nd harmonic signal is sufficient to identify acini, ducts, and blood vessels in the SMG by their morphological features. This setup can be used to observe transgenic fluorescent markers with cell type specific expression to identify specific cell subsets. Figure 2 shows a network ofCOL1A1-GFP reporter-expressing fibroblast-like cells of the connective tissue. Also shown are blood vessels, labeled by intravenous injection of a fluorescent dye coupled to high molecular weight dextran.The luminal compartment of the ducts can also be labeled (Figure 3) when the dye is injected retrogradually into the WD29.
Figure 1: Customized stage. (a and a*) two freely adjustable holders for cover slips; (b and b*) adjustable stereotactic ear bars; (c) a string, which can be tightened with a screw; (d) a freely adjustable metal heating ring; (e) raised area designated to place the torso of the mouse. Please click here to view a larger version of this figure.
Figure 2: Vasculature marker (red) and network of COL1A1-GFP expressing cells (green). Vasculature labeled by intravenous injection of Texas Red-dextran conjugate (MW 70 kDa). Fibrillar collagen in between two lobes is visualized as the 2nd harmonic signal (blue). Shown is a single z-projection of 15 stacked images with 47 µm depth. Scale bar = 40 µm. Please click here to view a larger version of this figure.
Figure 3: Luminal duct labeled after WD injection. Blood vessels labeled with 70 kDa-MW Texas Red-dextran (red), duct lumen labeled with cascade blue-dextran (white; 10 kDa MW) via retrograde injection through the WD and adoptively transferred GFP+ CD8+ T cells (green) in the SMG. Shown is a single z-projection of 27 stacked images with 50 µm depth. Scale bar = 40 µm. Please click here to view a larger version of this figure.
This protocol offers a straightforward approach for in vivo imaging of murine submandibular and sublingual salivary glands using upright non-linear microscopy often used in the field of immunology. The method can be adapted for the imaging of other exocrine glands in the head and neck region. For instance, our lab has performed imaging of the lacrimal gland in an analogous manner (not shown).
Removing the connective tissue around the SMG is the most critical step of this protocol, since accidental tissue damage can occur. Damage to the main blood supply of the SMG necessitates immediate termination of the experiment, since the SMG tissue will become hypoxic. If the SMG tissue itself is damaged, bleeding typically stops naturally by coagulation. However, it needs to be considered that the injury triggers sterile inflammation and proinflammatory cytokines release30, which could influence the experimental readout. A more common, but less critical complication during the microsurgery are leaks in the saline reservoir. To minimize the occurrence of leaks, always apply grease to dry, hairless skin. Patching a leak is usually futile: instead, remove all of the grease, clean the skin with a dry cotton swab, and reapply from scratch.
During imaging, a temperature close to the physiological body temperature must be maintained (approximately 37 °C for male and female C567BL/6 mice31). Since both hypo- and hyperthermia alter blood flow32,33, physiological function of the SMG cannot be assumed under those conditions. However, we observed no difference in blood flow or T cell motility during small (± 2 °C), transient (<5 min) deviations from optimal temperature.
Minimizing tissue movement is essential for good imaging quality. Different kinds of unwanted tissue motion can occur, and have typical causes that can undergo troubleshooting. Firstly, continuous xyz-drift is usually caused by the SMG slowly drifting back into its original position through attached connective tissue. This happens if the lower cover glass was installed too high, or the connective tissue around the SMG was not thoroughly disrupted. Secondly, slow and continuous z-drift may occur if the temperature is unstable. In this case carefully adjust the heating and wait for the temperature to be stable, approximately 15 min before imaging. Finally, pulsating tissue motion occurs when the saline chamber is not tightly sealed. Pressure builds up and releases in the saline chamber with every breathing cycle. In combination with leaks or air bubbles this leads to pulsating fluid flows, which translate directly to the SMG tissue. In this case, reassembly of both the cover glasses and saline chamber is the most reliable solution.
The maximum imaging depth depends on the excitation wavelength used and the fluorophores employed. As in virtually all intravital imaging setups, red-shifted dyes or fluorescent proteins that can be excited with long wavelengths (900-1,100 nm) allow for deeper tissue imaging than green dyes and shorter excitation wavelengths.
The SMG has been used as model organ for endo- and exocytosis and membrane remodeling3,27,28, infection34, autoimmunity11, and tumor biology12. It further permits many forms of manipulation: dyes, pharmacological compounds, and antibodies can be delivered to the gland via the blood stream. Alternatively, the SMG can be targeted directly via the cannulation of the WD, which has been used to deliver virus11, dextrans3, DNA25, dyes29, or pharmacological agents3.
The intravital imaging field continues to profit from an ever-growing toolbox of genetic and biochemical fluorescent labeling techniques. Advances in genetic engineering provide mouse lines and cell lines with sophisticated fluorescent markers. Recent examples include labeling of specific cell subsets (CD11c+ cells35) or dynamic cytoskeletal proteins (Lifeact36), photoconvertible markers for cell tracking 37, and real-time reporters of nuclear translocation (NFAT38) or calcium signaling39. Other applications may use fluorescent biochemical probes, which highlight specific subcellular compartments (such as the DNA binding 4',6-diamidino-2-phenylindole (DAPI) or lipophilic membrane dyes). Modern NLO microscopy offers many possible solutions to best answer the experimental questions. For instance, multiplexed two-photon imaging has been developed to strongly increase maximum acquisition speed40. Other techniques offer label-free multiphoton imaging, either by generating second or third harmonic signals15, or by measuring characteristic resonance vibration of molecules (coherent anti-Stokes Raman scattering)41. Thus, when practicable procedures for tissue exposure and stabilization are available, the limits of experimental design are mostly set by the experimenter's resources and creativity.
The authors have nothing to disclose.
This work was funded by Swiss National Foundation (SNF) project grant 31003A_135649, 31003A_153457 and 31003A_172994 (to JVS), and Leopoldina fellowship LPDS 2011-16 (to BS). This work benefitted from optical setups of the “Microscopy Imaging Center” (MIC) of the University of Bern.
Narketan 10 % (Ketamine) 20ml (100 mg/ml) | Vetoquinol | 3605877535982 | |
Rompun 2% (Xylazine) 25 ml (20 mg/ml) | Bayer | 680538150144 | |
Saline NaCl 0.9% | B. Braun | 3535789 | |
Prequillan 1% (Acepromazine) 10 ml (10 mg/ml) | Fatro | 6805671900029 | |
Electric shaver | Wahl | 9818L | or similar |
Hair removal cream | Veet | 4002448090656 | |
Durapore Surgical tape (2.5 cm x 9.1 m and 1.25 cm x 9.1 m) | Durapore (3M) | 1538-1 | |
Durapore Surgical tape (2.5 cm x 9.1 m and 1.25 cm x 9.1 m) | Durapore (3M) | 1538-0 | |
Super glue Ultra gel, instantaneous glue | Pattex, Henkel | 4015000415040 | |
Microscope cover glass slides 20 mm and 22 mm | Menzel-Gläser | 631-1343/ 631-1344 | |
Grease for laboratories 60 g glisseal N | Borer (VWR supplier) | DECO514215.00-CA15 | |
Surgical scissors | Fine Science Tools (F.S.T ) | 14090-09 | or similar |
Fine Forceps | Fine Science Tools (F.S.T ) | 11252-20 | or similar |
Cotton swab | Migros | 617027988254 | or similar |
Gauze Gazin 5 x 5 cm | Lohmann and Rauscher | 18500 | or similar |
Stereomicroscope | Leica | MZ16 | or similar |
Texas Red dextran 70kDa | Molecular Probes | D1864 | |
Cascade Blue dextran 10kDa | invitrogen | D1976 | |
Two-photon system | LaVision Biotec | TrimScope I and II | or similar |
XLUMPLANFL 20x/0.95 W objective | Olympus | n/a | or other water immersion objective |
Digital thermometer | Fluke | 95969077651 |