0448

Oxidative stress measured by in-vivo, longitudinal 1H MRS and ex-vivo ESR spectroscopy in a rat model of chronic Hepatic Encephalopathy
Katarzyna Pierzchala1, Veronika Rackayova1, Olivier Braissant2, Dario Sessa3, Stefanita Mitrea1, Andrzej Sienkiewicz4, Valérie A. McLin3, Rolf Gruetter1, and Cristina Cudalbu1

1Center for Biomedical Imaging, EPFL, Lausanne, Switzerland, 2Service of Biomedicine, CHUV, Lausanne, Switzerland, 3Swiss Center for Liver Disease in Children, Department of Pediatrics, HUG, Geneva, Switzerland, 4Laboratory of Physics of Complex Matter, EPFL, Lausanne, Switzerland

Synopsis

Oxidative stress is believed to play a role in the pathogenesis of chronic hepatic encephalopathy (CHE), however no in-vivo studies to assess the course of cerebral oxidative stress in CHE were performed. We investigated the oxidative stress in the hippocampus of a rat model of chronic HE using in-vivo and longitudinal 1H-MRS combined with ex-vivo ESR spectroscopy. Early changes in brain antioxidants (Asc, 4weeks post-BDL), concomitant with Gln and NH4+ changes were detected for the first time in-vivo. These results were confirmed by ESR spectroscopy suggesting that central oxidative stress is an early event in CHE.

PURPOSE

Reactive oxygen species (ROS) are small messenger molecules that are normal components of signal transduction cascades during physiological processes but are also involved in neurotoxicity and neurodegeneration. The redox homeostasis is maintained by the balance between ROS generation and elimination by antioxidants. ROS are critical for hippocampal long-term potentiation, a synaptic plasticity for learning and memory as well as in aging/disease-related impairment1. When in excess ROS lead to cellular dysfunction.

Oxidative stress is believed to play a role in the pathogenesis of chronic hepatic encephalopathy (CHE)2. CHE is accepted to occur in part from glutamine (Gln) accumulation secondary to impaired ammonium (NH4+) clearance by the diseased liver, and Gln accumulation leads to mitochondrial dysfunction. Impaired brain NH4+ detoxification together with Gln increase induces ROS generation in turn associated with astrocyte swelling3,4. To date there are no in-vivo studies to assess the course of cerebral oxidative stress in CHE.

The aim of this study was to investigate the oxidative stress in the hippocampus of a rat model of chronic HE using in-vivo and longitudinal 1H-MRS combined with ex-vivo ESR spectroscopy.

METHODS

In-vivo 1H-MRS was performed on a 9.4T-MR-system (Varian/Magnex-Scientific) using a home-built 14mm-diameter quadrature 1H-surface coil as a transceiver. The SPECIAL sequence5 was used (TE=2.8ms,TR=4sec;160averages). The hippocampus (2x2.8x2mm3) of Wistar male adult rats was scanned before bile-duct ligation (BDL-animal model of CHE) and after every 2weeks up to week8 (n=20). First and second-order shims were adjusted using FASTESTMAP. Concentrations of metabolites were calculated by LCModel using water as reference.

For ESR experiments the hippocampus was extracted at 6weeks post-BDL (n=6) and sham-surgery, weighed, sliced and transferred into whole RPMI1640 medium (Sigma) with 10mM CMH spin trap (Noxygen GmbH), incubated at 37oC. After each incubation time tissue suspension was transferred into quartz capillary tube. Experiments were carried out at RT using an ESP300E spectrometer (Bruker BioSpin GmbH), equipped with a TE102 cavity.

Extracted hippocampus was incubated in the dark with Nitro Blue Tetrazolium (1.6 mg/mL) in HBSS at 37oC for 60 min for histochemical detection of ROS species.

RESULTS AND DISCUSSION

As expected, the presence of chronic liver disease was confirmed by the early increase in plasma NH4+ and bilirubin (2weeks post-BDL). The characteristic neurochemical pattern of CHE was also observed: gradual increase of Gln (significant at week4 post-BDL),significant decrease in main osmolytes (week6 post-BDL) as an osmotic answer to Gln accumulation (Ins-30%, tCho-26%, Tau-8%, Cr -11%) and neurotransmitters decrease (Glu-11%) (Fig.1).

Inositol acts as a multifaceted player in cell signalling and has an important role in maintenance of normal brain function. Neurotransmitters like norepinephrine, dopamine, serotonin, acetylcholine and GABA rely on inositol to relay messages6-9. Alteration in Ins levels influences the calcium pump and in consequence might imbalance the ROS homeostasis10.

Oxidative stress is known to be an important feature of HE. Therefore, we focused on the evolution of two important brain antioxidants, Asc and GSH, using 1H-MRS. Asc showed a statistical significant decrease at week4 compared to week0 (-12%,p<0.05) and reached -15% at week8 (p<0.01). Asc decreased over time correlating mainly with plasma NH4+ increase (r=-0.37,p=0.002) but no correlation was observed with brain Gln (Fig.1), suggesting that the two may be directly linked in the pathogenesis of oxidative stress11 in the central nervous system of rats with CHE. We also observed a trend toward decreasing levels of GSH (-16% at 8weeks post-BDL).

The ESR CMH spin trapping was applied to evaluate ROS levels in hippocampus ex-vivo. CMH is an effective cell permeable and non-toxic spin probe for the quantification of slow released intracellular O2·-. Statistically significant increase of hippocampal ROS levels in BDL (p<0.01, Fig.2) corroborates the 1H-MRS findings of decreasing Asc (Fig.1).

NBT staining reflected increased ROS activity of the fimbria hippocampi (FH) in BDL rats (Fig.3). FH is made of myelinated axons providing the major conduit for subcortical connections, forming fimbria-fornix path12. A significant difference in hippocampus size and weight was observed between shams (120.8±20mg) and BDL animals (71.1±12.7mg)(p<0.01). Therefore, it is tempting to extrapolate that high ROS levels are associated with a loss of cholin-, GABA- and glutamatergic neurons as well as with behavioural deterioration.

CONCLUSIONS

The present study shows the potential to monitor oxidative stress in a rat model of CHE using in-vivo and longitudinal 1H-MRS combined with ex-vivo ESR spectroscopy. We detected, for the first time in-vivo, early changes in brain antioxidants (4weeks post-BDL), concomitant with Gln and NH4+ changes. These results were confirmed by ESR spectroscopy suggesting that central oxidative stress is an early event in CHE.

Acknowledgements

Supported by CIBM of the UNIL, UNIGE, HUG, CHUV, EPFL, the Leenaards and Jeantet Foundations and the SNSF project no 310030_173222/1.

References

1Lauren T. Knapp and Eric Klann, Journal of Neuroscience Research, 2002; 2Bosoi CR, Rose CF. Metab Brain Dis. 2013 Jun;28(2):175-8; 3Skowronska M. and Albrect J., Neurochemistry International, 2012; 4Lachman V. et al., Archives of Biochemistry and Biophysics, 2013; 5Mlynárik et al, Magn Reson Med 2006; 6Michael J. Berridge, Biochem. Soc. Trans. 2012; 7Boittin, Franc¸ois-Xavier et al., the American Physiological Society,1999; 8Joseph Levine, European Neuropsychopharmacology, 1997; 9Y Ohmori, K Kuriyama, Neurochem. Int., 1989; 10A.K.M. Shamsuddin and Guang-Yu Yang,Inositol & its Phosphates: Basic Science to Practical Applications, 2015; 11Rama Rao et al. 2003; Noremberg 2004; 12The Hippocampus Book, Oxford University Press, 2007; 13 Baehner RL, Boxer LA, Blood, 1976; 14 Armstrong JS et al., Int J Androl. 2002.

Figures

A)Longitudinal evolution of plasma total-bilirubin and ammonium; statistical significant levels between week0 and all the post-operative weeks are shown in black, while for week2 vs all the other post-operative weeks are shown in grey only for total bilirubin since the value at week0 was non-measurable. B)Longitudinal evolution of Gln, Ins, tCho, Tau, and Cr, from week0 (before-BDL) to 8weeks after BDL. C)Longitudinal evolution of Asc from week0 to 8weeks post-BDL. Pearson correlations: brain Gln vs. brain Asc and plasma ammonium vs. brain Asc. Data are presented as mean ±SD and statistical significance (One-way-Anova) is always presented compared to week 0:*p<0.05,**p<0.01,***p<0.001,****p<0.0001.

A)~50μL of tissue suspension transferred into 1.5mm ID and 1.8mm OD quartz capillary tubes (VitroCom, 25mm). B)Cell permeable and non-toxic spin probe for the quantification of slow released intracellular O2·- production and detection of ROS in-vitro and in-vivo. The spin adduct is resistant to reduction by vitamin C and thiols, therefore allows quantitative detection of intracellular O2·-. C)ESR signal time evolution of the intracellular spin-trap: BDL rat hippocampus. Focus on the middle line. D)ROS generation rate calculated from ESR kinetic plots (EPR signal levels versus the elapsed time), and corresponding p-value for this change, **p<0.01, presented as mean±SD.

Hippocampus and fimbria hippocampi (FH) stained with NBT at week 8 after BDL: superoxide anions detection by histochemical staining, which is based on the ability of cells to reduce NBT a water-soluble yellow nitro-substituted aromatic tetrazolium to a dark blue precipitate13,14. NBT specifically reacts with superoxide radicals and forms a purple / blue formazan precipitate. NBT staining indirectly reflects the ROS-generating activity of living cells, and therefore can help to determine the redox status of cells. NBT accumulation show directly increased ROS level in BDL FH.

Proc. Intl. Soc. Mag. Reson. Med. 26 (2018)
0448