Melville J. Phillips

Byler's disease: Fatal intrahepatic cholestasis

An infant with Bylers disease from the originally described Amish kindred was studied by the use of isotopic bile acids and electron microscopy to further delineate the hepatic defects in this recessively inherited fatal intrahepatic cholestatic disease. To the already documented abnormalities of transport of Bromsulphalein and conjugated bilirubin are added new features of abnormal bile acid metabolism, which demonstrate a defect in hepatic excretion of conjugated bile acid across the canalicular membrane. The accumulation of bile acids (especially lithocholic acid) may possibly cause canalicular damage. Electron microscopic changes in the liver consisted of previously described altered microvilli and thickened canalicular membrane, and a new finding of coarse particulate bile. This male infant and his father have an abnormal small Y chromosome similar to those observed in other male patients with Bylers disease; this finding is unrelated to the intrahepatic cholestasis.

Glycogen depletion in the newborn rat liver an electron microscopic and electron histochemical study

In the rat, the hepatic glycogen store extant at birth is almost completely mobilized in the first 12 hours of life. The close-meshed reticulum of smooth membranes (SER) which has been associated with glycogenolysis in the adult rat liver cell is virtually absent throughout the period of glycogen mobilization in the newborn. This finding is strongly supported by glycogen digestion procedures since the hyaloplasm of the digested specimens completely lacks any trace of a network of smooth membranes. We have concluded that the SER plays no role in the mobilization of glycogen in these liver cells. Many cytolysomes containing glycogen (glycogenosomes) were found 6 hours following birth. Acid phosphatase was localized within the glycogenosomes establishing their lysosomal nature. Digestion of the hyaloplasmic glycogen sometimes failed to remove the glycogen within glycogenosomes. We have concluded that lysosomal enzymes contribute, at least in some measure, to glycogen catabolism in these cells.

The FGL2-FcγRIIB pathway: A novel mechanism leading to immunosuppression

Fibrinogen‐like protein 2 (FGL2) is a multifunctional protein, which has been implicated in the pathogenesis of allograft and xenograft rejection. Previously, FGL2 was shown to inhibit maturation of BM‐derived DC and T‐cell proliferation. The mechanism of the immunosuppressive activity of FGL2 remains poorly elucidated. Here, we focus on identification of FGL2‐specific receptor(s) and their ability to modulate APC activity and allograft survival. Using flow cytometry and surface plasmon resonance analysis, we show that FGL2 binds specifically to Fc gamma receptor (FcγR)IIB and FcγRIII receptors, which are expressed on the surface of APC, including B lymphocytes, macrophages and DC. Antibody to FcγRIIB and FcγRIII, or deficiency of these receptors, abrogated FGL2 binding. FGL2 inhibited the maturation of BMDC from FcγRIIB+/+ mice but not from FcγRIIB−/− mice and induced apoptosis in the FcγRIIB+ mouse B‐cell line (A20) but not the A20IIA1.6 cell line that does not express FcγRIIB. Recombinant FGL2 infused into FcγRIIB+/+ (C57BL/6J, H‐2b) mice but not FcγRIIB−/− mice inhibited rejection of fully mismatched BALB/cJ (H‐2d) skin allografts. The identification of specific receptor binding has important implications for the pathogenesis of immune‐mediated disease and suggests a potential for targeted FGL2 therapy.

The novel immunoregulatory molecule FGL2: A potential biomarker for severity of chronic hepatitis C virus infection

BACKGROUND & AIMS This report describes the use of a novel sensitive and specific ELISA for the measurement of human fibrinogen-like protein 2 (FGL2/fibroleukin), a novel effector of natural regulatory T (Treg) cells, to predict the course of chronic hepatitis C viral infection (HCV). METHODS Plasma levels of FGL2 were measured in HCV patients and compared to healthy controls and to patients with alcoholic liver disease. RESULTS FGL2 levels were significantly higher in HCV patients (84.3+/-89.1 ng/ml, n=80) compared to healthy controls (36.4+/-21.9 ng/ml, n=30, p<0.001), to a subset of patients who cleared HCV following anti-viral treatment (16.6+/-19.7 ng/ml, n=32, p<0.001), and to patients with inactive alcoholic liver disease (18.8+/-17.4 ng/ml, n=24, p<0.001). Among HCV patients, plasma levels of FGL2 correlated significantly with the stage of fibrosis (p=0.001) and were significantly higher in patients with cirrhosis (164.1+121.8 ng/ml, n=60) compared to non-cirrhotics (57.7+/-52.8 ng/ml, n=20, p=0.001). Genotype 1 patients had significantly higher levels of FGL2 (98.1+/-100.3 ng/ml, n=60) compared to patients with genotype 2/3 (41.5+/-38.6 ng/ml, n=20, p=0.0008). Patients with genotype 2/3 had FGL2 levels similar to healthy controls (41.5+/-38.6 vs. 36.41+/-21.9 ng/ml, p=ns). Infiltrating lymphocytes in liver biopsies of HCV patients were positive for either FGL2 or FoxP3 (a marker of Treg cells) or expressed both markers. CONCLUSIONS This report documents the development of a sensitive ELISA for measurement of plasma levels of FGL2 an effector Treg cells, which correlates with the severity of HCV infection.

FGL2/Fibroleukin mediates hepatic reperfusion injury by induction of sinusoidal endothelial cell and hepatocyte apoptosis in mice

BACKGROUND & AIMS Sinusoidal endothelial cell (SEC) and hepatocyte death are early, TNF-α mediated events in ischemia and reperfusion of the liver (I/Rp). We previously reported that TNF-α induced liver injury is dependent on Fibrinogen like protein 2 (FGL2/Fibroleukin) and showed that FGL2 binding to its receptor, FcγRIIB, results in lymphocyte apoptosis. In this study we examine whether I/Rp is induced by specific binding of FGL2 to FcγRIIB expressed on SEC. METHODS Hepatic ischemia and reperfusion was induced in wild type (WT) mice and in mice with deletion or inhibition of FGL2 and FcRIIB. Liver injury was determined by AST release, necrosis and animal death. Apoptosis was evaluated with caspase 3 and TUNEL staining. RESULTS FGL2 deletion or inhibition resulted in decreased liver injury as determined by a marked reduction in both levels of AST and ALT and hepatocyte necrosis. Caspase 3 staining of SEC (12% vs. 75%) and hepatocytes (12% vs. 45%) as well as TUNEL staining of SEC (13% vs. 60%, p=0.02) and hepatocytes (18% vs. 70%, p=0.03), markers of apoptosis, were lower in Fgl2(-/-) compared to WT mice. In vitro incubation of SEC with FGL2 induced apoptosis of SEC from WT mice, but not FcγRIIB(-/-) mice. Deletion of FcγRIIB fully protected mice against SEC and hepatocyte death in vivo. Survival of mice deficient in either Fgl2(-/-) (80%) or FcγRIIB(-/-) (100%) was markedly increased compared to WT mice (10%) which were subjected to 75min of total hepatic ischemia (p=0.001). CONCLUSIONS FGL2 binding to the FcγRIIB receptor expressed on SEC is a critical event in the initiation of the hepatic reperfusion injury cascade through induction of SEC and hepatocyte death.

Hepatic ultrastructural changes in acute fructose overload

Ultrastructural changes in rat hepatocytes begin within 5 minutes of fructose infusion. The earliest detectable change is a rarefaction of the hyaloplasm which rapidly increases, and is apparently due to overhydration or edema. It is not accompanied by swelling of mitochondria or alteration in the density of the matrix. The smooth endoplasmic reticulum forms large nodular masses near the periphery of the cell; presumably by redistribution of the agranular reticulum rather than by new formation of membrane. The Golgi zones become extremely small and simplified. The cisternae of the rough endoplasmic reticulum show disarray, and segmental loss of ribosomes, and contribute to the formation of membranous whorls. It is suggested that many of these changes are due to the sudden drop in ATP which is known to occur in acute fructose loading.

The prothrombinase activity of FGL2 contributes to the pathogenesis of experimental arthritis

Objective: Fibrin deposition is integral to the pathogenesis of collagen-induced arthritis (CIA), an experimental model of rheumatoid arthritis (RA). Membrane-associated fibrinogen-like protein 2 (mFGL2), a novel inducible prothrombinase, generates fibrin by an alternate pathway and has been reported to be involved in the pathogenesis of a number of immune-mediated diseases. We hypothesized that expression of mFGL2 in inflamed synovium contributes to the fibrin deposition and subsequent inflammation in arthritis. Methods: DBA/1 mice were immunized with 100 µg bovine collagen type II (CII) emulsified in complete Freunds adjuvant (CFA) followed by lipopolysaccharide (LPS) injection. Expression of mFGL2 prothrombinase in association with fibrin deposition was examined in mice with CIA and CD200-treated mice following induction of CIA. To directly assess the contribution of mFGL2, fgl2−/− mice were injected with antibody to CII (anti-CII). Results: Levels of fgl2 mRNA transcripts and mFGL2 protein were markedly up-regulated in joints of mice that developed CIA. Fibrin deposition was prominent within the synovial lining and articular joint space associated with expression of mFGL2. Inhibition of CIA by the immunosuppressant CD200 was associated with decreased expression of fgl2 mRNA and mFGL2 protein and absence of fibrin deposition. Following injection of anti-CII, all fgl2+/+ mice developed severe arthritis with clinical and histological manifestations characteristic of RA, whereas fgl2−/− mice failed to develop any clinical manifestation or histological evidence of arthritis. Conclusions: This study demonstrates that the prothrombinase activity of mFGL2 contributes to the pathogenesis of experimental arthritis. These studies may have therapeutic implications for patients with RA.

Genetic Determinants of Mouse Hepatitis Virus Strain 1 Pneumovirulence

ABSTRACT We report here investigation into the genetic basis of mouse hepatitis virus strain 1 (MHV-1) pneumovirulence. Sequencing of the 3′ one-third of the MHV-1 genome demonstrated that the genetic organization of MHV-1 was similar to that of other strains of MHV. The hemagglutinin esterase (HE) protein was truncated, and reverse transcription-PCR (RT-PCR) studies confirmed previous work that suggested that the MHV-1 HE is a pseudogene. Targeted recombination was used to select chimeric viruses containing either the MHV-1 S gene or genes encoding all of the MHV-1 structural proteins, on an MHV-A59 background. Challenge studies in mice demonstrated that expression of the MHV-1 S gene within the MHV-A59 background (rA59/SMHV-1) increased the pneumovirulence of MHV-A59, and mice infected with this recombinant virus developed pulmonary lesions that were similar to those observed with MHV-1, although rA59/SMHV-1 was significantly less virulent. Chimeras containing all of the MHV-1 structural genes on an MHV-A59 background were able to reproduce the severe acute respiratory syndrome (SARS)-like pathology observed with MHV-1 and reproducibly increased pneumovirulence relative to rA59/SMHV-1, but were still much less virulent than MHV-1. These data suggest that important determinants of pneumopathogenicity are contained within the 3′ one-third of the MHV-1 genome, but additional important virulence factors must be encoded in the genome upstream of the S gene. The severity of the pulmonary lesions observed correlates better with elevated levels of inflammatory cytokines than with viral replication in the lungs, suggesting that pulmonary disease has an important immunological component.

The bile canalicular network in vitro

Bile canaliculus-rich fractions of rat liver plasma membranes have been shown by routine and three-dimensional scanning electron microscopy to be comprised not of single discrete canaliculi but of relatively intact portions of the bile canalicular network.

Electron microscopy of the liver.

There is little doubt that standardization of electron microscopic technic, study of selected liver biopsies, and correlation of the findings with light microscopic, biochemical and clinical observations will add much to our understanding of liver disease. The structure of human liver cells in health and disease is reviewed.