Search

Heme oxygenase (HO) catalyzes the oxidative cleavage of the alpha-mesocarbon of Fe-protoporphyrin-IX yielding equimolar amounts of biliverdin-IXa, iron, and carbon monoxide. The HO-system consists of two isoenzymes, namely HO-2 and the inducible isoform HO-1, also referred to as heat shock protein (hsp) 32. Although both parenchymal and non-parenchymal liver cells participate in heme metabolism, the expression pattern of the isoenzymes in normal and stress exposed liver is unknown. To study this, rats underwent either endotoxin (lipopolysaccharide [LPS]) challenge, hemorrhagic hypotension, glutathione (GSH) depletion, or cobalt chloride injection, all known to provoke oxidative stress. HO-2 messenger RNA (mRNA) and protein were constitutively expressed in hepatocytes, Kupffer/endothelial-, and stellate (Ito-) cell enriched fractions. Although both non-parenchymal cell fractions expressed HO-1 transcripts, HO-1 immunoreactive protein was restricted to Kupffer cells in the normal liver. In contrast to HO-2, a significant increase in HO-1 on the whole organ level was noted by hemorrhagic hypotension, GSH depletion, and cobalt chloride injection. However, the distinct stress models led to a strikingly different cell-type specific and sublobular expression pattern of HO-1 gene expression. HO-1 was inducible in sinusoidal lining cells (hemorrhagic hypotension, LPS challenge), in periportal (cobalt chloride), or pericentral (GSH depletion, hemorrhagic hypotension) hepatocytes. The blockade of protein translation before hemorrhage by cycloheximide reduced upregulation of HO-1/hsp32 mRNA significantly (65.4% reduction, P < .05), whereas the inducibility of hsp70 transcript was maintained. In addition to transcriptional regulation, HO-1 seems to be subject to posttranscriptional control in particular in non-parenchymal cells.

In contrast to the anticipation that in sepsis granulocyte colony-stimulating factor (G-CSF) would overactivate the nonspecific immune system by recruiting and priming leukocytes with consequent aggravation of inflammatory tissue lesions, recombinant (r) G-CSF pretreatment was protective in various experimental non-neutropenic models of inflammation. The mechanisms of protection, however, are not fully understood. Using intravital fluorescence microscopy, we show that rG-CSF enhances leukocyte endothelial cell interaction within the microvasculature of normal rat livers, whereas rG-CSF pretreatment of animals exposed to lipopolysaccharide (LPS) attenuates the LPS-induced leukocytic response, including stasis in sinusoids as well as rolling and adherence in postsinusoidal venules with subsequent tissue infiltration. Moreover, rG-CSF, which did not affect Kupffer cell activity in normal rat livers, reduced the immediate activation of Kupffer cells on LPS exposure, as indicated in vivo by the delayed adherence/phagocytosis of intra-arterially administered latex particles associated with attenuation of proinflammatory cytokine release (tumor necrosis factor α and interleukin-6). Finally, rG-CSF reduced LPS-induced nutritive perfusion failure and hepatocellular excretory dys function. This study provides evidence for a distinct, possibly tumor necrosis factor α-dependent modulation of LPS-induced cellular response within the liver by rG-CSF, thereby achieving protection against microcirculatory perfusion failure and hepatic dysfunction. J.

Heme oxygenase (HO)-derived carbon monoxide (CO) may contribute to vascular control through elevation of guanosine 3',5'-cyclic monophosphate. In the present study, we investigated the functional significance of expression of the isoenzyme HO-1 (heat-shock protein 32) in liver after hemorrhage/resuscitation (H/R) in rats anesthetized with pentobarbital sodium. An increase of mRNA levels for HO-1 was observed at 3 h after resuscitation, followed by induction of the protein at 6 h in pericentral hepatocytes and sinusoidal lining cells. Concomitantly, lower portal resistance was observed in H/R (0.33 +/- 0.060 mmHg.ml-1.min) compared with control rats (0.47 +/- 0.035 mmHg.ml-1.min). Blockade of the HO-CO pathway by tin protoporphyrin-IX (SnPP-IX) led to a transient increase in portal pressure with no effect on portal low in controls, whereas an increase in pressure and a decrease in flow contributed to the sustained increase in portal resistance after H/R. These results indicate that HO contributes to maintenance of hepatic perfusion in vivo under stressful conditions, suggesting a functional link between stress response and vascular control in portal circulation.