**3. Effects of Ornithinephenylacetate on glutaminase activity in bile duct rats: Inflammation and ammonia**

In cirrhotic patients it has been shown that the effects of hyperammonemia are synergistic with inflammation (Shawcross, Davies et al. 2004). The effects on cell swelling by cytokines in ammonia-sensitized cultured astrocytes has also been shown (Rama Rao, Jayakumar et al. 2010). However, the mechanisms by which ammonia produces brain swelling are still subject of much investigation. Although the effects on inflammatory processes have been found to contribute to the formation of cerebral edema, it is not clear whether ammonia promotes inflammation or both are independent factors. Inflammatory pathways identified as contributing to the edema include cyclo-oxygenase, nitric oxide (NO)/cyclic guanosine monophosphate (cGMP) signaling and cytokine release (Andrade, Lucena et al. 2005; Montoliu, Piedrafita et al. 2009; Montoliu, Rodrigo et al. 2010).

transcriptional activity. In Figure 1 is shown that patients with long microsatellite showed

0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 **Follow-up (months)**

Fig. 1. Actuarial curve showing patients free of bouts of overt hepatic encephalopathy according to the microsatellite in the promoter region. Patients with long microsatellite

**3. Effects of Ornithinephenylacetate on glutaminase activity in bile duct rats:** 

In cirrhotic patients it has been shown that the effects of hyperammonemia are synergistic with inflammation (Shawcross, Davies et al. 2004). The effects on cell swelling by cytokines in ammonia-sensitized cultured astrocytes has also been shown (Rama Rao, Jayakumar et al. 2010). However, the mechanisms by which ammonia produces brain swelling are still subject of much investigation. Although the effects on inflammatory processes have been found to contribute to the formation of cerebral edema, it is not clear whether ammonia promotes inflammation or both are independent factors. Inflammatory pathways identified as contributing to the edema include cyclo-oxygenase, nitric oxide (NO)/cyclic guanosine monophosphate (cGMP) signaling and cytokine release (Andrade, Lucena et al. 2005;

showed higher risk of overt hepatic encephalopathy (log-rank: 7.74; p<0.01)

Montoliu, Piedrafita et al. 2009; Montoliu, Rodrigo et al. 2010).

higher risk of overt hepatic encephalopathy.

1.0

0.8

0.6

0.4

**% Patients free of overt HE**

0.2

0.0

**Inflammation and ammonia** 

Hyperammonemia could increase blood-brain-barrier permeability to systemic cytokines. It is also possible that several factors associated with the systemic inflammatory response syndrome could modulate brain dysfunction induced by hyperammonaemia. These processes, together with a genetic factor, may help to explain the differences that sometimes exist between lower ammonia levels and observed brain impairment in some patients. It has been shown that the presence of HE grade 3/4 correlates better with inflammation than with ammonia plasma levels (Shawcross, Sharifi et al. 2010), though extracellular brain ammonia levels may be significantly higher. One recent study showed that in a cirrhosis animal model in which plasma and brain cytokines were markedly elevated following administration of lipopolysaccharide (LPS), pre-treatment with OP prevented increased levels of TNFα and IL-6 (trend) in plasma and in brain observed in the control group (Wright G 2010). Moreover, OP reduced LPS induced development of pre-coma/coma and worsening of brain edema. It is well-known that the transcription of NFkB directly increases proinflammatory cytokines and leads to induction of nitric oxide synthase. (Li and Verma 2002). OP reduced iNOS and NFkB expressions in the cortical brain region of cirrhotic animals, indicating that ammonia reduction may modulate neuroinflammation. In cirrhosis a paradox exists between reduced intrahepatic NO generation and excess NO in the splanchnic circulation. Splanchnic vasodilatation leads to vasoconstriction of numerous vascular beds, including the liver, kidneys, and has significant effects on the brain. Asymmetric dimethylarginine (ADMA) is an endogenous inhibitor of eNOS (endothelial nitric oxide synthase), the levels of which are increased in liver failure (Leiper, Nandi et al. 2007; Mookerjee, Malaki et al. 2007). It has been shown that treatment of cirrhotic rats with OP resulted in restoration of the NOS pathways which may have a direct effect on cerebral perfusion (Balasubramaniyan, Wright et al. 2011).

Physical symptoms of HE and minimal HE have been detected by motor-evoked potentials (MEP) which examines the function of signal transmission along the nerve, which is perturbed by low grade brain edema. Similar disturbances have been found in patients with cirrhosis using magnetic resonance (MR), with signs compatible with low-grade edema along the corticospinal tract. These abnormalities were related to functional impairment detected by transcranial magnetic stimulation and were found to be reversed after liver transplantation. Recently the assessment of MEP in awake rats has been validated to monitor HE in animal models of liver failure (induced by portocaval anastomosis-PCA) and precipitated HE (simulated gastrointestinal bleed-GB). These models have been utilized to test the efficacy of OP, demonstrating that OP treatment prevents the neurophysiological abnormalities induced by a GB insult in PCA animals. Administration of OP over differing time periods (between 3 hours and 3 days) as a pretreatment prevents the decrease in the amplitude and increase in MEP latency at 6 hours post GB (Oria M, Romero-Gimenez J et al. 2011).

In preliminary studies, it has been shown that the combination of ornithine with phenylacetate to treat hyperammonaemia in cirrhosis is effective in animal models. Administration of OP results in increased conversion of glutamate to glutamine by stimulation of GS activity in the muscle with the subsequent excretion of phenylacetylglutamine in the urine. GA has been found to contribute to hyperammonaemia in cirrhosis and in hyperammonaemia animal models (Romero-Gomez, Grande et al. 2004; Romero-Gomez, Jover et al. 2006).

In this novel approach to target the altered interorgan ammonia metabolism in liver failure, OP utilizes the activity of GS to trap ammonia as glutamine, then phenylacetate facilitates its excretion as phenylacetylglutamine (Davies, Wright et al. 2009; Ytrebo, Kristiansen et al. 2009). The effectiveness of this approach with OP has been confirmed in animal models of cirrhosis and ALF. The reduction (≈50%) of plasma ammonia was associated with (a) an improvement in grade of HE in cirrhotic patients and (b) a reduction in ICP in acute liver failure. OP treatment significantly reduced ammonia concentrations, which was associated with a reduction in brain water and an increase in myoinositol levels, indicating an improvement in brain metabolism (Jalan, Wright et al. 2007; Davies 2009; Ytrebo LM 2009).

In a devascularised pig model of ALF, the rise in arterial ammonia was attenuated with OP which was accompanied by a significant decrease in extracellular brain ammonia and prevention of intracranial hypertension (Ytrebo, Sen et al. 2006).

In our studies we included twenty-five male Sprague-Dawley rats: 4 sham operated, and 21 BDL. 5 BDL's received OP (5 days, IP 0.6 g/kg), 5 BDL's received ornithine (5 days, IP 0.6 g/kg), 5 BDL's received phenylacetate (5 days, IP 0.6 g/kg) and 6 received saline (IP). We measured plasma levels for: ammonia and standard biochemical markers. Expressions of GS, GA and ornithine amino transferase (OAT) were determined by Western-blot (expressed as a % of sham values) and enzyme activity was determined by end-point methods in liver, kidney, gut, muscle and lung. We found that plasma ammonia was decreased in BDL-OP rats vs. BDL-saline (58.97±6.02 vs. 106.2±20.56 µmol/L;P<0.05). BDL-OP rats showed increased GS expression in liver (66% BDL-OP vs. 55% BDL-saline; P < 0.01) and showed further increased levels in the muscle (153% BDL-OP vs.142% BDL-saline). OP ameliorates the BDL related increases in glutaminase expression (124%vs.163%; P<0.05) and activity (0.45±0.16mIU/mg protein BDL-OP vs. 1.14±0.046mIU/mg protein BDL-saline; P<0.01) in gut. We demonstrated that this prevention is due to effect of ornithine in glutaminase activity (0.46±0.17mIU/mg protein BDL-O vs. BDL-saline; P<0.05) and not to phenylacetate. OP treatment in BDL rats increased the conversion of glutamate to glutamine by stimulation of GS in the muscle and also resulted in normalization of glutaminase expression and activity in the gut, indicating that OP effectively restricts the production of *in vivo* ammonia in a cirrhotic. Mechanism of action of OP on the metabolism of ammonia is shown in the Figure 2.
