1.

Introduction

One of the major hepatobiliary functions is the uptake, processing, and excretion of a variety of endogenous or exogenous organic anions by the hepatocytes. The anions are obtained from the sinusoids and conjugated with glutathione, glucuronate, or sulfate in hepatocytes, and then excreted by an ATP-dependent mechanism into the bile canaliculi through the apical membrane of hepatocytes.¹ Conventional methods to investigate these processes usually involved the biochemical analyses of the contents in the liver, bile, or other body fluids such as blood or urine.² Real-time imaging of hepatobiliary excretion has been approached on ex-vivo liver tissues by laser confocal microscopy.³ However, the ex-vivo model cannot be used to investigate the in-vivo dynamics of hepatic activities. Therefore, the establishment of an intravital imaging system aimed at following these functional processes may be of tremendous value in hepatology.

In this work, we report the design and application of an intravital hepatic imaging chamber on the mouse for the in-vivo investigation of the hepatobiliary excretory function of 6-CFDA.

2.

Methods

2.1.

Design of the Intravital Hepatic Imaging Chamber

Our hepatic imaging chamber is composed of two doughnut-shaped titanium rings: the outer and inner lids [Figs. 1a and 1b ]. The outer and inner diameters of the outer lid are 12.5 and $6.5\mathrm{mm}$ , respectively, with a thickness of $1\mathrm{mm}$ . For the inner lid, the outer and inner diameters are 11.5 and $6.5\mathrm{mm}$ , respectively, and it has a thickness of $0.8\mathrm{mm}$ . Both lids are threaded and can be fastened together by simple screwing.

Fig. 1

Illustrations showing the structure of the hepatic imaging chamber and its anatomic position after installation on the mouse abdominal wall. (a) Horizontal and (b) oblique views of the chamber apparatus. (c) A photograph showing a mouse with the device installed on its upper abdominal wall. (d) A diagram depicting the anatomic relationships of the installed device with skin, peritoneum, and the underlying liver.

2.3.

Accommodation of the Mouse onto the Microscopic Stage

The positioning of the mouse installed with the hepatic imaging chamber and the subsequent imaging procedures are illustrated in Fig. 2 . The mouse was kept in the supine position with the hepatic imaging chamber fitted into the U-shaped groove of a steel plate [Fig. 2a], and the plate was slid toward jaw of the mouse and lodged the liver window device to the terminus of the groove [Fig. 2b]. The mouse with the steel plate was then placed on the microscopic stage of an inverted microscope (TE2000, Nikon, Japan) by screwing the plate onto the microscopic stage [Fig. 2c]. In this manner, the mouse can be firmly attached for multiphoton imaging purposes. A hot pack between the gauze was put on top of the mouse for warming purposes [Fig. 2d].

Fig. 2

Accommodation of the mouse on the microscope stage. Fixing of the mouse with its hepatic imaging chamber into the groove of a steel plate was shown in (a), (b), and (c). (d) Fixing the plate onto the microscope stage by screws. (e) A brief schematic of the imaging light path.

3.

Results

Figures 3 and 4 shows the representative multiphoton images acquired using the tissue adhesive Histoacryl. The depth for imaging hepatobiliary function was set at approximately $30\mathrm{\mu }\mathrm{m}$ below the disappearance of the SHG signals from the capsule. Upon injection of 6-CFDA through the jugular vein, the sinusoids already demarcated by the red fluorescence. In addition, the scattered intensely fluorescent spots (5 to $15\mathrm{\mu }\mathrm{m}$ ) on the margins of the sinusoids were visible [Fig. 3a]. These structures are identified as the stellate cells that contain vitamin A emitting strong autofluorescence.⁴ Following 6-CFDA injection, green fluorescence starts to increase in the hepatic cords between the sinusoids and peaks at the 8-min point [Figs. 3b, 3c, 3d]. The green fluorescence intensity then starts to decrease and almost reaches the background level at approximately 50-min post-6-CFDA injection [Fig. 3f].

Fig. 3

Representative intravital multiphoton hepatic images acquired using Histoacryl. (a) through (f) are the time-lapsed multiphoton images acquired approximately $30\mathrm{\mu }\mathrm{m}$ below the capsule. Time in minutes indicates period post-6-CFDA injection. Small arrow in (a) indicates the presence of stellate cells (blue), while large arrow indicates sinusoid positions (red). Arrow in (b) indicates hepatocytes in cords emitting CF fluorescence (green), and arrows in (c), (d), and (e) indicate CF fluorescence in bile canaliculi.

Fig. 4

Representative intravital multiphoton hepatic images acquired using Histoacryl. Depth-dependent multiphoton images acquired approximately $50\min$ after 6-CDFA injection. The stellate cell (small arrow) and sinusoid (large arrow) are clearly visible in (c).

Soon after finishing the time-course observations at approximately 50-min post-6-CFDA injection, we imaged the same microscopic field at different depths of 5, 20, 30, and $50\mathrm{\mu }\mathrm{m}$ [Figs. 4a, 4b, 4c, 4d]. The green fluorescence visible in the bile canaculi at earlier time points is no longer apparent. This observation is consistent with the fact that the uptake, processing, and excretion of 6-CFDA is now complete.

The kinetics of the mean green fluorescence intensity of the liver is shown in Fig. 5 . Also shown are the intensity profiles (6-CFDA jugular vein injection) acquired using the adhesives of 4011 and 406. As the figure demonstrates, although the intensity profiles obtained using the three adhesives show different temporal dependence, hepatic metabolism of 6-CFDA is mostly complete at the 50-min mark. Since we only applied the adhesives at the chamber edge, we feel that the effects of glue on hepatic metabolism are minimal. In addition, since the intensity profile of 406 decayed the fastest, the biocompatibility of the glues appear not to be a factor in affecting the hepatic metabolic activities of 6-CFDA. Therefore, we feel that the difference in measured intensity profiles may be due to variations in the individual mouse metabolism and the variations in 6-CFDA injection procedures.

Fig. 5

Measured normalized fluorescence intensity profiles using the glues of Histoacryl, 4011 (for medical device, Henkel Loctite), and 406 (instant adhesive, Henkel, Loctite).

4.

Discussion

Multiphoton microscopy offers several advantages in the intravital imaging of living tissues. It is superior to videomicroscopy⁵ in its high depth-discrimination nature and also in providing simultaneous multicolor observation of tissue specimens. Multiphoton imaging is also superior to confocal microscopy due to the increased imaging penetration depth and markedly reduced phototoxicity.⁶ In addition, SHG microscopy can also be used to detect collagen fibers in the tissues without the addition of fluorophores.⁷ Intravital multiphoton microscopy has been used in the investigations of renal physiological and pathological processes,⁸ but to the best of our knowledge, it has never been applied in the hepatology field. In this work, we have successfully demonstrated the monitoring of hepatic metabolic activities by the use of multiphoton microscopy and an intravital hepatic imaging chamber.

During the course of this work, we found that the vibration of liver tissues caused by respiration and heartbeat can be minimized by gluing the edges of the inner lid of the hepatic imaging chamber to the underlying liver with tissue adhesive. The adhesive-free regions of the liver were used for imaging purposes. The preadministered rhodamine dextran stays in circulation for as long as $5\mathrm{h}$ and thus the sinusoids and venules in the liver are well demarcated by its red fluorescence. 6-CFDA can be taken up by the hepatocytes and then be hydrolyzed by esterase into carboxyfluorescein (CF), which would emit at $517\mathrm{nm}$ .⁹ After additional hepatocyte processing, CF is then excreted into bile.¹⁰ Our data clearly showed the sequential illumination by CF of hepatocytes and then bile canaliculi, indicating the hepatobiliary excretion of this organic anion. After the microscopic examination, the animals might, carrying the hepatic imaging chembron their abdomen, live as usual, needing no restriction on activities.

These results support the use of our methodology to be a powerful technique to investigate the hepatobiliary excretory functions in diseases, such as extrahepatic and intrahepatic cholestasis. In addition, hepatic metabolism of chemicals and drugs over an extended period of several days may be investigated with our approach. The application of imaging over a large tissue area may also open a way to observe acute or chronic liver injuries, or follow targeted cell migration and proliferation in complex liver environments.