Liver sinusoidal endothelial cells represents an important blood clearance system in pigs

Abstract

BACKGROUND: Numerous studies in rats and a few other mammalian species, including man, have shown that the sinusoidal cells constitute an important part of liver function. In the pig, however, which is frequently used in studies on liver transplantation and liver failure models, our knowledge about the function of hepatic sinusoidal cells is scarce. We have explored the scavenger function of pig liver sinusoidal endothelial cells (LSEC), a cell type that in other mammals performs vital elimination of an array of waste macromolecules from the circulation. RESULTS: 125I-macromolecules known to be cleared in the rat via the scavenger and mannose receptors were rapidly removed from the pig circulation, 50% of the injected dose being removed within the first 2-5 min following injection. Fluorescently labeled microbeads (2 &mgr;m in diameter) used to probe phagocytosis accumulated in Kupffer cells only, whereas fluorescently labeled soluble macromolecular ligands for the mannose and scavenger receptors were sequestered only by LSEC. Desmin-positive stellate cells accumulated no probes. Isolation of liver cells using collagenase perfusion through the portal vein, followed by various centrifugation protocols to separate the different liver cell populations yielded 280 x 107 (range 50-890 x 107) sinusoidal cells per liver (weight of liver 237.1 g (sd 43.6)). Use of specific anti-Kupffer cell- and anti-desmin antibodies, combined with endocytosis of fluorescently labeled macromolecular soluble ligands indicated that the LSEC fraction contained 62 x 107 (sd 12 x 107) purified LSEC. Cultured LSEC avidly endocytosed ligands for the mannose and scavenger receptors. CONCLUSIONS: We show here for the first time that pig LSEC, similar to what has been found earlier in rat LSEC, represent an effective scavenger system for removal of macromolecular waste products from the circulation.

Figure 1

Clearance kinetics. Approximately 100 × 10⁶ cpm of ¹²⁵I-tyramine cellobiose-formaldehyde-treated serum albumin (TC-FSA) and ¹²⁵I-α-mannosidase were injected intravenously. Radioactivity in the blood sample collected immediately after injection was taken as 100%. Blood samples were collected every minute during the first 10 minutes, then every 5 minutes for one hour. (Open boxes: ¹²⁵I-TC-FSA; n = 2, closed boxes: ¹²⁵I-α-mannosidase; n = 1).

Figure 2

Anatomical distribution. The animals used in the blood clearance studies (Fig. 1) were analyzed for anatomical distribution of radioactivity 1 h after injection. More than 90% of the injected doses were recovered in the organs listed. Results are expressed as percent total radioactivity recovered. (Grey bars: ¹²⁵I-TC-FSA; n = 2, white bars: ¹²⁵I-α-mannosidase; n = 1).

Figure 3

Fluorescence micrographs of liver section. Following intravenous administration of fluorescently labeled substances, sections were prepared as described in the Methods section. A heterogeneous distribution of yellow fluorescence from TRITC-labeled monodisperse polymer particles (MDPP) phagocytosed by Kupffer cells was located mainly in the periportal region of the liver acinus (arrows) (A). Green fluorescence along the lining of the liver sinusoids identifies endocytosed FITC-formaldehyde-treated serum albumin (FSA) by liver sinusoidal endothelial cells (LSEC), while the localization of phagocytosed MDPP is shown by arrows (B). Uptake of FITC-FSA (arrowheads) and MDPP (arrow) is shown more clearly at higher magnification in C. (Scale bars; A: 80 μm, B: 20 μm, C: 8 μm).

Figure 4

Uptake of monodisperse polymer particles (MDPP) in Kupffer cells (KC). Following intravenous administration of fluorescently labeled substances, sections were prepared as described in the Methods section for transmission electron microscopy. MDPP are located intracellularly in Kupffer cells, as judged by their characteristic phagocytosis of the particles (A). Hepatocytes (Hep) contain numerous mitochondria. The cells that contain fat vacuoles (FV) may represent stellate cells (SC). To distinguish between vacuoles containing fat and phagocytosed MDPP, sections were immunolabeled with monoclonal anti-mouse TRITC-conjugate. Gold particles are located in the periphery of MDPP where the TRITC-molecules are attached (B). (Scale bars; A: 2 μm, B: 500 nm).

Figure 5

Stellate cells (SC) and liver sinusoidal endothelial cells (LSEC). Following intravenous administration of fluorescently labeled substances, sections were prepared as described in the Methods section for transmission electron microscopy. Ultrathin sections were immunodouble labeled to visualize both FITC-labeled formaldehyde-treated serum albumin (FSA) in LSEC and desmin in SC. Figures B and C are higher magnification of segments of figure A. Cells lining the sinusoids (A) are LSEC as judged by the localization of small gold particles (5 nm, small arrow) in organelles taken as lysosomes (B). The cell containing large fatty vacuoles (FV) and large gold particles (10 nm, large arrow), was judged as a stellate cell (SC) (C). (Scale bars; A: 1 μm, B: 200 nm, C: 500 nm).

Figure 6

Fluorescence micrographs of cultured liver sinusoidal endothelial cells (LSEC). Cultures were prepared as described in the Methods section. The cultures were fixed in 4% paraformaldehyde, after 6 h of incubation. FITC-labeled formaldehyde-treated serum albumin (FSA) and TRITC-labeled monodisperse polymer particles (MDPP) were administered intravenously prior to isolation of liver cells. Fluorescent microscopy reveals a homogeneous LSEC culture contaminated by a few cells with TRITC-MDPP and lipid containing vacuoles. The green fluorescence from endocytosed FITC-FSA demonstrates that most cells are LSEC, and that the probe is localized in cytoplasmic vacuoles (A), whereas the yellow fluorescence from phagocytosed TRITC-MDPP identifies Kupffer cells (arrows) (B). Autofluorescence from vitamin A identifies stellate cells (arrows) (C). (Scale bars; 20 μm).

Figure 7

Fluorescent micrographs of cultured stellate cells and Kupffer cells. Cultures were prepared as described in Methods. The cultures were fixed in 4% paraformaldehyde, after 1 h of incubation. Micrographs of cultured stellate cells stained with monoclonal anti-desmin antibody (A) and cultured Kupffer cells stained with monoclonal anti-pig macrophage antibody (B). (Scale bars; 20 μm).

Figure 8

Specificity of endocytosis of ¹²⁵I-formaldehyde-treated serum albumin (FSA) in cultured liver sinusoidal endothelial cells (LSEC) (grey and white bars), and ¹²⁵I-asialo-orusomucoid protein (ASOR) in cultured hepatocytes (black and hatched bars). Monolayer cultures were incubated for 2 hrs, at 37°C, with trace amounts of labeled ligand alone (control) or together with excess amounts of unlabeled FSA (100 μg·mL^-1) or galactose (50 mmol·L^-1). The presence of unlabeled FSA inhibited effectively the endocytosis of ¹²⁵I-FSA in LSEC, while galactose showed no such inhibitory effect. Galactose had an inhibitory effect on endocytosis of ¹²⁵I-ASOR in hepatocytes, whereas unlabeled FSA showed no such inhibitory effect. Results, given as percent of control, are the means of triplicate experiments. Grey and white bars: 100% corresponds to 12.7% of added cpm, black and hatched bars: 100% corresponds to 14.6% of added cpm. White and hatched areas of bars represent % degraded ligand. Grey and black areas of bars represent % cell-associated ligand.