HAL Id: hal-01257127https://hal.archives-ouvertes.fr/hal-01257127

Submitted on 18 Jan 2016

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

Model-guided identification of a therapeutic strategy toreduce hyperammonemia in liver diseases

Ahmed Ghallab, Geraldine Celliere, Sebastian Henkel, Dominik Driesch,Stefan Hoehme, Ute Hofmann, Sebastian Zellmer, Patricio Godoy, Agapios

Sachinidis, Meinolf Blaszkewicz, et al.

To cite this version:Ahmed Ghallab, Geraldine Celliere, Sebastian Henkel, Dominik Driesch, Stefan Hoehme, et al.. Model-guided identification of a therapeutic strategy to reduce hyperammonemia in liver diseases. Journalof Hepatology, Elsevier, 2015, 64 (4), pp.860-871. �10.1016/j.jhep.2015.11.018�. �hal-01257127�

Our reference: JHEPAT 5901 P-authorquery-v13

AUTHOR QUERY FORM

s to:

Journal: JHEPAT Please e-mail your responses and any correction

E-mail: proofcorrections.esnl@elsevier.sps.co.in

) orlace

‘Q’

Article Number: 5901

Dear Author,

Please check your proof carefully and mark all corrections at the appropriate place in the proof (e.g., by using on-screen annotation in the PDF filecompile them in a separate list. Note: if you opt to annotate the file with software other than Adobe Reader then please also highlight the appropriate pin the PDF file. To ensure fast publication of your paper please return your corrections within 48 hours.

For correction or revision of any artwork, please consult http://www.elsevier.com/artworkinstructions.

Any queries or remarks that have arisen during the processing of your manuscript are listed below and highlighted by flags in the proof. Click on the

link to go to the location in the proof.

Location inarticle

Query / Remark: click on the Q link to goPlease insert your reply or correction at the corresponding line in the proof

Q1 Your article is registered as a regular item and is being processed for inclusion in a regular issue of the journal.If this is NOT correct and your article belongs to a Special Issue/Collection please contact j.bakker@elsevier.com immediately prior to returning your corrections.

Q2 The author names have been tagged as given names and surnames (surnames are highlighted in teal color).Please confirm if they have been identified correctly.

Q3 The country name has been inserted for the affiliation `6'. Please check, and correct if necessary.

Q4 The affiliations `7' and `11' were identical, the latter has been removed, and hence the affiliation link the author`Rolf Gebhardt' has been changed, accordingly. Please check, and correct if necessary.

Q5 One or more sponsor names may have been edited to a standard format that enables better searching andidentification of your article. Please check and correct if necessary.

Q6 The country names of the Grant Sponsors are provided below. Please check and correct if necessary. `BMBF' -`Germany'.

Thank you for your assistance.

Please check this box if you have nocorrections to make to the PDF file

Graphical abstract

pp xxx–xxxe hyperammonemia in liver diseases

JHEPAT 5901 No. of Pages 1, Model 5G

8 January 2016

Model-guided identification of a therapeutic strategy to reduc

* Ahmed Ghallab , Géraldine Cellière, Sebastian G. Henkel, Dominik Driesch,

Stefan Hoehme, Ute Hofmann, Sebastian Zellmer, Patricio Godoy, Agapios Sachinidis,Meinolf Blaszkewicz, Raymond Reif, Rosemarie Marchan, Lars Kuepfer, Dieter Häussinger,Dirk Drasdo, Rolf Gebhardt, Jan G. Hengstler *

1

1

2

3 of4 m

5 eba6 llm7 M8 rdt

9 ctors10 2D Med11 ocqu12 4 ioinfo13 Tuebi14 Un d Cen15 tems16 isea

1718

1920 Ba21 m22 ul23 el24 M25 us26 m27 ni28 ge29 ov30 bl31 R32 su33 m34 at35 tio36 (G37 fr38 am

39te

KeLivReNo⇑EnGeRe6740E-He

Abamalp

Q1

Q2

Q3

Q4

Research Article

JHEPAT 5901 No. of Pages 13

8 January 2016

PH

Model-guided identificationreduce hyperammone

Ahmed Ghallab1,2,⇑, Géraldine Cellière3,y, SStefan Hoehme5, Ute Hofmann6, Sebastian Ze

Meinolf Blaszkewicz1, Raymond Reif1, RosemarieDirk Drasdo3,5,�, Rolf Gebha

1Leibniz Research Centre for Working Environment and Human Faepartment of Forensic Medicine and Toxicology, Faculty of VeterinaryRecherche en Informatique et en Automatique (INRIA), INRIA Paris-RBioControl Jena GmbH, Jena, Germany; 5Interdisciplinary Centre for B

Fischer-Bosch Institute of Clinical Pharmacology and University ofiversity of Leipzig, Leipzig, Germany; 8Institute of Neurophysiology anRobert-Koch-Str. 39, 50931 Cologne, Germany; 9Computational Sys

10Clinic for Gastroenterology, Hepatology and Infectious D

ckground & Aims: Recently, spatial-temporal/metabolic

athematical models have been established that allow the sim-ation of metabolic processes in tissues. We applied these mod-

40th41in42In43N44co451546Co47th48fo49�50by 515253

54In

55Re56ge57[158by59co60tio61pr62Th63og64m

s to decipher ammonia detoxification mechanisms in the liver.ethods: An integrated metabolic-spatial-temporal model wased to generate hypotheses of ammonia metabolism. Predictedechanisms were validated using time-resolved analyses oftrogen metabolism, activity analyses, immunostaining andne expression after induction of liver damage in mice. More-er, blood from the portal vein, liver vein and mixed venousood was analyzed in a time dependent manner.esults:Modeling revealed an underestimation of ammonia con-mption after liver damage when only the currently establishedechanisms of ammonia detoxification were simulated. By iter-ive cycles of modeling and experiments, the reductive amida-n of alpha-ketoglutarate (a-KG) via glutamate dehydrogenaseDH) was identified as the lacking component. GDH is releasedom damaged hepatocytes into the blood where it consumesmonia to generate glutamate, thereby providing systemic pro-

ywords: Systems biology; Spatio-temporal model; Ammonia; Liver damage;er regeneration.ceived 23 March 2015; received in revised form 15 November 2015; accepted 16vember 2015Corresponding authors. Addresses: Leibniz Research Centre for Workingvironment and Human Factors, IfADo – Ardeystr. 67, D-44139 Dortmund,rmany. Tel.: +49 02311084 356; fax: +49 02311084 403 (A. Ghallab) or Leibniz

65at

66re67te68m69[170w71am72co73(G

search Centre for Working Environment and Human Factors, IfADo – Ardeystr., D-44139 Dortmund, Germany. Tel.: +49 02311084 349; fax: +49 02311084

3 (J.G. Hengstler).mail addresses: ghallab@ifado.de (A. Ghallab), hengstler@ifado.de (J.G.ngstler).y These authors contributed equally to the work.� These authors share co-senior authorship.breviations: CCl4, carbon tetrachloride; GDH, glutamate dehydrogenase; AOA,inooxy acetate; ALT, alanine transaminase; AST, aspartate transaminase; a-KG,ha-ketoglutarate; PDAC, 2,6-pyridinedicarboxylic acid.

Journal of Hepatology 201

lease cite this article in press as: Ghallab A et al. Model-guided identification ofepatol (2016), http://dx.doi.org/10.1016/j.jhep.2015.11.018

a therapeutic strategy toia in liver diseases

stian G. Henkel4,y, Dominik Driesch4,er7, Patricio Godoy1, Agapios Sachinidis8,archan1, Lars Kuepfer9, Dieter Häussinger10,7,�, Jan G. Hengstler1,⇑,�

at the Technical University Dortmund, Dortmund, Germany;icine, South Valley University, Qena, Egypt; 3Institute National deencourt & Sorbonne Universités UPMC Univ Paris 6, LJLL, Francermatics, University of Leipzig, Leipzig, Germany; 6Dr. Margaretengen, Germany; 7Institute of Biochemistry, Faculty of Medicine,ter for Molecular Medicine Cologne (CMMC), University of Cologne,Biology, Bayer Technology Services GmbH, Leverkusen, Germany;ses, Heinrich-Heine-University, Düsseldorf, Germany

ction against hyperammonemia. This mechanism was exploitederapeutically in a mouse model of hyperammonemia byjecting GDH together with optimized doses of cofactors.travenous injection of GDH (720 U/kg), a-KG (280 mg/kg) andADPH (180 mg/kg) reduced the elevated blood ammoniancentrations (>200 lM) to levels close to normal within onlymin.nclusion: If successfully translated to patients the GDH-basederapy might provide a less aggressive therapeutic alternativer patients with severe hyperammonemia.2015 European Association for the Study of the Liver. PublishedElsevier B.V. All rights reserved.

troduction

cent developments have strongly improved our capability tonerate information at multiple spatial and temporal scales,2]. However, research on disease pathogenesis is hamperedthe difficulty to understand the orchestration of individual

mponents. Here, mathematical models help to formalize rela-ns between components, simulate their interplay, and to studyocesses that are too complex to be understood intuitively [1].is is particularly important when studying the pathophysiol-y of metabolic liver diseases, where due to zonation differentetabolic processes take place in pericentral and periportal hep-ocytes [3]. To be able to investigate such complex processes wecently established a technique of integrated metabolic spatial-mporal modeling (IM) [4]. These IM integrate conventionaletabolic models into spatial-temporal models of the liver lobule,4,5]. The present study was motivated by the IM predictions,hich proposed that the conventional mechanisms wheremonia is metabolized by urea cycle enzymes in the periportalmpartments of the liver lobules and by glutamine synthetaseS) reaction in the pericentral compartments (Supplementary

6 vol. xxx j xxx–xxx

a therapeutic strategy to reduce hyperammonemia in liver diseases. J

74 s75 l76 e77 d78 n79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100101102103104105106107108109110111112113114115116117118119120121122123124125

126

127

128

129

130

131

132d133g134r135n136-

Research Article

JHEPAT 5901 No. of Pages 13

8 January 2016

Fig. 1) failed to explain the experimental findings [4]. The IM waapplied to an experimental scenario, where the entire pericentraand a part of the periportal compartment of the liver lobules werdestroyed by a single high dose of the hepatotoxic compouncarbon tetrachloride (CCl4). This leads to compromised nitroge

metabolism and hyperammonemia. In the present study, we per-formed a series of new experiments accompanied by simulationswith novel models to explore the mechanism responsible for theobserved discrepancy. Experimentally, the time-resolved analysisof metabolites and metabolic activities after CCl4 intoxicationoffers good conditions to study ammonia detoxification and pos-sible compensatory mechanisms during the damage and regener-ation process. Time-resolved analysis of metabolites wasperformed in the portal vein and heart blood, representing the‘liver inflow’, and in the liver vein as ‘liver outflow’. These analy-ses allowed a precise experimental validation of model predic-tions. Finally, iterative cycles of modeling and experimentalvalidation allowed the identification of a so far unrecognizedmechanism of ammonia detoxification. Importantly, this mecha-nism could be exploited therapeutically to reduce elevated bloodammonia concentrations close to normal levels by intravenousinjection of glutamate dehydrogenase (GDH; 720 U/kg) and itscofactors alpha-ketoglutarate (a-KG; 280 mg/kg) as well asNADPH (180 mg/kg). This example illustrates how concrete ther-apies can be derived by model guided experimental strategies.

Materials and methods

A detailed description of materials and methods is provided in the Supplemen-tary materials. Male C57BL/6N 10–12 weeks old mice were used (Charles River,Sulzfeld, Germany). Acute liver damage was induced by intraperitoneal injectionof 1.6 g/kg CCl4, unless other doses are indicated. Blood was taken from miceunder anesthesia from the portal and hepatic veins, as well as the right heartchamber, and plasma was separated. Liver tissue samples were collected fromdefined anatomical positions for histopathology, immunohistochemistry, enzymeactivity assays, gene array and q-RT-PCR analyses. The dead cell area was quan-tified in hematoxylin and eosin stained tissue sections using Cell^M software(Olympus, Hamburg, Germany). Whole-genome analysis of gene expression inmouse liver tissue was performed in control as well as after CCl4 intoxicationwith Affymetrix gene arrays. The latter techniques are described fully in the Sup-plementary materials and methods. The analysis of ammonia and furthermetabolites was performed using commercially available kits. Concentrationsof amino acids and organic acids in liver tissue were measured in duplicate usingGC-MS. GS, GDH and transaminases activity assays were performed photometri-cally as described in the Supplementary materials and methods. NADP+ andNADPH were analyzed by LC-MS. Mouse hepatocytes were isolated by a two-step EGTA/collagenase perfusion technique and either used directly in suspen-sion or cultivated in collagen sandwiches (Supplementary materials and meth-ods). For the mathematical modeling of ammonia and the related metabolitesthe integrated metabolic, spatio-temporal model was applied [4,5]. In addition,the IM was replaced by a set of novel models that include further reactionsand the blood compartment of the liver (Supplementary materials and methods).Statistical analysis was done with SPSS software as described in the Supplemen-tary materials.

Results

An integrated spatial-temporal-metabolic model suggests a so farunrecognized mechanism of ammonia detoxification

The detoxification process in healthy, damaged and regeneratinglivers was simulated using a recently established integratedmetabolic IM [4]. To compare the simulated metabolite concen-

Please cite this article in press as: Ghallab A et al. Model-guided identificatioHepatol (2016), http://dx.doi.org/10.1016/j.jhep.2015.11.018

2 Journal of Hepatology 2

trations with the in vivo situation, an experiment was performein which blood was collected from the portal vein (representin85% of the ‘liver inflow’), the heart (representing 15% of the ‘liveinflow’), and the hepatic vein (representing the ‘liver outflow’) ia time-resolved manner after CCl4 injection (Fig. 1A; Supplemen

137tary Fig. 2). The result shows that ammonia is detoxified during138its passage through the liver as illustrated by the difference in139ammonia concentrations between the portal vein and the hepatic140vein in the control mice (Fig. 1B). This detoxification process is141compromised after liver damage, particularly on days 1 and 2.142Surprisingly, the IM model predicted higher ammonia concentra-143tions than those experimentally observed, particularly on day 1144(Fig. 1C; see the video in the Supplementary data). Analyses of145heart blood demonstrate the contribution of the extrahepatic146compartment, which includes brain, muscles, kidneys and blood,147to ammonia detoxification between days 1 and 4 after the induc-148tion of liver damage. However, this extrahepatic contribution is149small compared to detoxification by the liver (Supplementary150Figs. 2–8). In addition to the time-resolved study, similar151experiments were also performed in a dose dependent manner152on day 1 after CCl4 administration when the discrepancy between153simulated and measured ammonia was maximal. For this pur-154pose, doses ranging between 10.9 and 1600 mg/kg CCl4 were155tested, resulting in a concentration dependent increase in the156dead cell area, with only the highest dose causing damage to157the entire CYP2E1 positive pericentral region of the liver lobule158(Fig. 1D; Supplementary Fig. 9A, B). Destruction of the GS positive159area occurred in with doses ranging between 38.1 and 132.4 mg/160kg (Fig. 1D, E; Supplementary Fig. 9C); also CPS1 showed a161dose dependent decrease (Supplementary Fig. 9C) leading to162compromised ammonia metabolism (Supplementary Fig. 10).163Using the IM [4], we also observed a discrepancy between the164predicted and measured ammonia in the dose dependent study165(Fig. 1F).166To find an explanation for this discrepancy, we performed167time-resolved gene array analysis of mouse liver tissue after168CCl4 intoxication (Fig. 2A). Fuzzy clustering identified seven gene169clusters which reflected time dependent gene expression alter-170ations [6]. Clusters 4 and 6 contained genes whose expression171was transiently repressed at early time points after CCl4 intoxica-172tion (Fig. 2B). Further bioinformatics analyses revealed an over173representation of nitrogen/ammonia metabolism KEGG and Gene174ontology terms of genes in cluster 4 (Fig. 2C, D). Genes relevant175for ammonia metabolism were further studied by qRT-PCR,176immunostaining and activity assays. GS is the key enzyme for177ammonia detoxification in the pericentral compartment. RNA178levels of GS started to decrease as early as 6 h after CCl4 injection,179it was at its lowest between days 1 and 4, before finally recover-180ing to initial levels between days 6 and 30 (Fig. 2E). A similar181time-dependent curve was obtained for GS activity although182the decrease occurred slightly later than that of RNA with very183low levels between days 2 and 4 (Fig. 2E). The pattern and inten-184sity of GS immunostaining was found to be comparable to GS185activity (Fig. 2F). In addition, ornithine aminotransferase (OAT),186an enzyme exclusively localized in GS positive pericentral hepa-187tocytes that provides additional glutamate for fixing ammonia188[7], decreased to almost undetectable levels with a delayed189recovery (Supplementary Fig. 3A). The key enzymes of the190periportal compartment, CPS1, ASS1, ASL and arginase1 were191similarly analyzed in the same tissue (Supplementary Figs. 3B192and 4). Extending the IM [4], with time-dependent

n of a therapeutic strategy to reduce hyperammonemia in liver diseases. J

016 vol. xxx j xxx–xxx

193 en194 an195 (S196 va197 de

198Ac199by

200Fu201de

.0

.5

.2

.3

.1

.4

.0

.5

.2

.3

.1

.4

132.4

JOURNAL OF HEPATOLOGY

PH

JHEPAT 5901 No. of Pages 13

8 January 2016

zyme concentrations (model 1), did not remove the discrep-cy between model predictions and experimental dataupplementary Fig. 11), indicating that our model lacks a rele-nt, but so far unrecognized mechanism of ammoniatoxification.

Heart

Analyzed parameters• ammonia• urea• glutamine• glutamate• glucose• lactate• pyruvate• alanine• α-ketoglutarate• arginine• other amino acids• ALT, AST

Site

s of

blo

od c

olle

ctio

n

Hepaticvein

Portalvein

A

0

0

Amm

onia

(mM

)

0

0

0

0

B

0

0

Amm

onia

(mM

)

0

0

0

0C

0 h 12 h 24 h0.2 mM

2 d 4 d 6 d

0.0 mM

D Control 10.9 mg/kg 38.1 mg/kg

Gro

ssly

H

&E

CYP

2E1

GS

E

00.0 10

.938

.113

2.446

0.0

1600

.0

600

1000

**

*

GS

activ

ity(m

U/m

g pr

otei

n)

Dose of CCl4 (mg/kg)

200

800

400

0.0

Amm

onia

(mM

)

0.2

0.3

0.1

0.4F

lease cite this article in press as: Ghallab A et al. Model-guided identification ofepatol (2016), http://dx.doi.org/10.1016/j.jhep.2015.11.018

Journal of Hepatology 2016

ute liver damage provides systemic protection against ammoniaGDH release

rther evidence that an unrecognized mechanism of ammoniatoxification exists arose from metabolic analyses performed

0 h 1 h 6 h 12 h 1 d 2 d 3 d 4 d 6 d 12 d 30 d

**

** * *

Portal vein

Time after CCl4 administration

Hepatic veinHeart

*****

***

***

*

****

meas, inmeas, outsim, out

0 h 1 h 6 h 12 h 1 d 2 d 3 d 4 d 6 d 12 d 30 dTime after CCl4 administration

mg/kg 460 mg/kg 1600 mg/kg

**

**

***meas, inmeas, outsim, out

0.0 10.9

38.1

132.4

460.0

1600

.0

Dose of CCl4 (mg/kg)

a therapeutic strategy to reduce hyperammonemia in liver diseases. J

vol. xxx j xxx–xxx 3

202 using plasma from mice after CCl4 injection (Fig. 3A). Most of the203 analyzed factors in plasma (urea, glutamine, glucose, lactate,204 pyruvate, alanine, arginine and other amino acids: Supplemen-205 tary Figs. 4–6) were within the expected concentration ranges,206 a-KG, which dramatically decreased between 12 h and day 2207 (Fig. 3A). This decrease was accompanied by an almost concur-208 rent increase in glutamate levels, which persisted longer than209 the drop in a-KG. One potential explanation is the delayed recov-210 ery of GS, which uses glutamate and ammonia to form glutamine211 (Fig. 2E, F). The decrease in a-KG (and the increase in glutamate)212 was also accompanied by increased GDH activity in plasma,213 because GDH is released from damaged hepatocytes (Fig. 3A).214 The present observations suggest that GDH released from the215 damaged hepatocytes into the blood catalyzes, at least tran-216 siently, a reaction that consumes ammonia to produce glutamate217 (Fig. 3D). To test this hypothesis, we collected plasma from mice218 on day 1 after CCl4 injection. Addition of a-KG alone was suffi-219 cient to slightly but significantly decrease blood ammonia con-220 centrations (Fig. 3B). This decrease was enhanced by further221 adding NADPH and particularly GDH; whereas the GDH inhibitor,222 PDAC completely antagonized the effect. To test also higher223 ammonia concentrations typically observed in patients with sev-224 ere pre-coma hyperammonemia, 600 lM ammonia was added to225 plasma collected on day 1 after CCl4 administration. Under these226 conditions, a-KG also reduced ammonia and increased glutamate227 concentrations (Fig. 3C; Supplementary Fig. 12A). Together, these228 experiments suggest that a GDH reaction consuming ammonia in229 blood takes place when GDH is released from acutely damaged230 livers (Fig. 3D).

231 Validation of the ‘GDH-driven ammonia consumption’ in hepatocytes

232 The experiments described above suggest that high ammonia233 concentrations in plasma leads to a ‘reverse’ GDH reaction, which234 consumes rather than produces ammonia. To test whether this235 ‘GDH-driven ammonia consumption’ occurs not only in plasma236 but also in cells, we used an in vitro system with primary mouse237 hepatocytes incubated with ammonia in suspension (Fig. 4).238 PDAC was used to inhibit GDH (Fig. 4A) in order to determine239 its influence on ammonia metabolism. In hepatocytes isolated240 from control mice, unphysiologically high ammonia concentra-241 tions (2 mM) were required until PDAC caused a significant242 increase of ammonia levels in the suspension buffer (Fig. 4B).243 However, when hepatocytes from mice 24 h after CCl4 intoxica-244 tion were used, PDAC treatment increased ammonia concentra-245 tions in the suspension buffer, even with 0.5 mM ammonia.246 Furthermore, in the absence of ammonia, hepatocytes secreted247 a small but statistically significant amount of ammonia into the248 buffer. Similarly, glutamate production was reduced by PDAC,

249an effect that was also stronger in hepatocytes isolated from250CCl4-exposed mice (Fig. 4C), which corresponds to the reverse251GDH reaction proposed in Fig. 3D (right panel). CCl4 destroys252the pericentral hepatocytes (Fig. 1D), which explains the reduced253glutamine generation by GS (Fig. 4D) and compromises urea cycle254enzymes (Supplementary Fig. 3B, C), which explains the reduced255urea production (Fig. 4E). Similar experiments were also per-256formed with cultivated (instead of suspended) hepatocytes from257untreated mice. The results demonstrate that inhibition of GDH258at high ammonia concentrations increases ammonia-induced259cytotoxicity (Supplementary Fig. 12B). These results show that260the catalytic direction of GDH reverses a clearly becomes ammo-261nia consuming also in hepatocytes in order to compensate the262compromised metabolism by urea cycle enzymes and GS after263intoxication.264Further evidence emerges from simulations with a set of265novel models 1–4 (Supplementary Fig. 11). If a reversible GDH266reaction was integrated into the hepatocyte compartment267(Fig. 5A; Supplementary Fig. 11), the discrepancy between268in vivo measured and simulated ammonia concentrations269(Fig. 1C) completely disappeared (Fig. 5B). The quantitative270agreement was obtained even without considering the blood271compartment of the liver, suggesting that after CCl4-induced272damage, the ammonia consumption catalyzed by GDH in273the hepatocytes represents the missing ammonia sink predicted274by [4].

275Therapy of hyperammonemia based on the reverse GDH reaction

276The above described ammonia consumption catalyzed by the277GDH reaction (Figs. 3B–D and 4) and the aforementioned278decrease in plasma a-KG levels (Fig. 3A) prompted us to test279whether supplementation of a-KG in mice helps to detoxify280ammonia. Therefore, mice received a hepatotoxic dose of CCl4281(1.6 g/kg) and 24 h later a-KG (280 mg/kg) was injected into282the tail vein. Blood was collected immediately before as well as28315, 30 and 60 min after injection of a-KG. A decrease in plasma284ammonia concentrations by 31, 40 and 43% was observed 15,28530 and 60 min after a-KG injection, respectively (Fig. 6A). Gluta-286mate increased after 15 min and decreased again after longer287periods probably due to the consumption by further metabolism.288a-KG transiently increased in plasma after injection and then289rapidly decreased. Analysis of GDH activity demonstrated that290the experiment was performed under conditions of high plasma291activity. In control mice, injection of a-KG did not alter blood292concentrations of ammonia or glutamate (Fig. 6B). In addition,293plasma a-KG levels were lower in CCl4-treated mice compared294to the control mice, suggesting increased consumption in mice295with damaged livers.

Fig. 1. Evidence for a so far unrecognized mechanism of ammonia detoxification. (A) Experimental design. (B) Ammonia concentrations in the portal vein, hepatic veinand heart. ⁄p <0.05 compared to the corresponding controls (0 h). (C) Integrated metabolic spatio-temporal model using the technique described by [4] (video in theSupplementary data). Predicted ammonia concentrations in the liver outflow are higher compared to the experimental data. ⁄⁄⁄p <0.001, ⁄⁄p <0.01 and ⁄p <0.05 compared tothe measured ammonia output. (D) Dose dependent experiment (10.9 to 1600 mg/kg CCl4 24 h after administration) showing macroscopic alterations with a spottedpattern at 132.4 mg/kg and higher doses, corresponding to the central necrotic lesion in hematoxylin/eosin staining, scale bars: 100 lm. Destruction of the pericentralCYP2E1 positive region which begins at 132.4 mg/kg with central necrosis still surrounded by CYP2E1 positive surviving hepatocytes; the entire CYP2E1 positive region wasdestroyed at the highest dose of 1600 mg/kg. The GS positive region was destroyed only at 132.4 mg/kg and higher doses, which corresponds to the decrease in GS activity(E), scale bars: 200 lm. ⁄p <0.05 when compared to the control group (0). (F) Comparison of analyzed and simulated ammonia concentrations in the liver vein for theexperiment in (D); meas, in: analyzed concentrations in the portal vein (representing 85% of the liver inflow) and heart blood (representing 15% of the liver inflow); meas.out: analyzed concentrations in the liver vein; sim. out: simulated concentrations in the liver vein. Data are mean values and SD of three mice per time point and dose ofCCl4. ⁄⁄⁄p <0.001 and ⁄⁄p <0.01 compared to the measured ammonia output.

Research Article

Please cite this article in press as: Ghallab A et al. Model-guided identification of a therapeutic strategy to reduce hyperammonemia in liver diseases. JHepatol (2016), http://dx.doi.org/10.1016/j.jhep.2015.11.018

4 Journal of Hepatology 2016 vol. xxx j xxx–xxx

JHEPAT 5901 No. of Pages 13

8 January 2016

0 1 2 3 4 6 30

CCl4 (1.6 g/kg) A B

0 10 15

0.0

-1.0

-0.5

0.5

1.0

5

Cluster 4310 genes

Fold

cha

nge

Time after CCl4 administration (days)

KEGG msilobatem negortin Time after CCl4 noitartsinimda lobmyS Name Healthy

liver Hour 2 Hour 8 Day 1 Day 2 Day 4 Day 6 Day 8 Day 16

Car1 1 1 esardyhna cinobraC -1.12 -1.78 -2.42 -1.85 -1.11 1.1 1.15 1.02

Car3 1 3 esardyhna cinobraC -2.94 -4.51 -32.55 -4.47 -1.33 -1.74 -1.71 -1.41

Car5a Carbonic anhydrase 5a, lairdnohcotim 1 -1.64 -4 -4.44 -1.96 1.1 -1.17 1.02 -1.1

Cth Cystathionase (cystathionine )esayl- 1 -1.28 -1.95 -2.59 -2.03 1.35 -1.08 -1.04 -1.09

Gls2 2 esanimatulG )lairdnohcotim ,revil( 1 -1.65 -2.44 -3.71 -2.13 -1.06 -1.26 -1.01 -1.14

Glul Glutamate-ammonia ligase )esatehtnys enimatulg( 1 1.13 -1.84 -5.82 -4.85 -2.36 -1.02 -1.27 -1.08

Hal 1 esayl ainomma eniditsiH -1.71 -3.09 -3.86 -2.11 1.05 -1.37 1.14 1

C

D ssecorp .batem aeru/elcyc aeru OG Time after CCl4 noitartsinimda

lobmyS Name Healthy liver Hour 2 Hour 8 Day 1 Day 2 Day 4 Day 6 Day 8 Day 16

Aldh6a1 Aldehyde dehydrogenase 1A ylimafbus ,6 ylimaf 1 -1.15 -1.63 -2.33 -2.09 -1.24 1.02 1.02 -1.03

Dpyd Dihydropyrimidine esanegordyhed 1 -1.28 -1.71 -2.63 -2.15 -1.27 -1.08 1.07 -1.02

Dpys 1 esanidimirypordyhiD -1.26 -2.91 -3.35 -1.87 -1.08 -1.01 -1.05 -1.12

Asl 1 esayl etaniccusoninigrA 1.06 1.5 -2.12 -1.74 -1.13 -1.13 -1.05 -1.09

Cebpa CCAAT/enhancer binding ahpla ,)PBE/C( nietorp 1 -1.25 -1.35 -2.07 -1.4 1.06 1.14 1.05 -1.07

Otc Ornithine transcarbamylase 1 -1.22 -1.89 -3.26 -2.14 1.06 -1.08 1.01 -1.01

E

0.0

0.6

1.2

*

*

* * **

*

*0.9

0.3

0 h 1 h 6 h 12 h 1 d 2 d 3 d 4 d 6 d 12 d 30 d

GS

RN

A (2

-ΔΔc

t )

Time after CCl4 administration

0

150

250

* * *

*

*

200

50

100

0 h 1 h 6 h 12 h 1 d 2 d 3 d 4 d 6 d 12 d 30 d

GS

activ

ity (U

/live

r)

F

12 h

4 d 30 d

Control 1 d

6 d

Time (days)

Fig. 2. Spatio-temporal alterations of ammonia metabolizing enzymes after CCl4 intoxication. (A) Experimental design. (B) Time dependent changes of gene expressionin fuzzy cluster 4 from [6]. The dots correspond to the average of the mean scaled values for all 310 genes, between their respective maximal and minimal expression levelsat each time point, using healthy liver (time 0) as reference. Error bars indicate standard error. (C) Changes in expression of genes associated to the KEGG terms ammonia/nitrogen metabolism (Gene Ontology [GO] ID 910) as revealed by KEGG pathways enrichment analysis in fuzzy cluster 4 (p = 2.36 � 10�7). (D) Changes in the expression ofgenes associated to the GO terms ‘urea cycle/urea metabolic process’ (Gene Ontology ID 0000050 and 0019627 respectively) as revealed by GO enrichment analysis in fuzzycluster 4 (p = 3.83 � 10�4). In C and D, the values indicate fold of expression over healthy liver at each time point after CCl4 administration, and correspond to the average of5 independent biological replicates. Time course of GS RNA levels, GS activity (E) and immunostaining (F), Scale bars: 200 lm. ⁄p <0.05 when compared to the control group(0 h).

JOURNAL OF HEPATOLOGY

Please cite this article in press as: Ghallab A et al. Model-guided identification of a therapeutic strategy to reduce hyperammonemia in liver diseases. JHepatol (2016), http://dx.doi.org/10.1016/j.jhep.2015.11.018

Journal of Hepatology 2016 vol. xxx j xxx–xxx 5

JHEPAT 5901 No. of Pages 13

8 January 2016

00 h 1 h 6 h 12 h 1 d 2 d 3 d 4 d 6 d 12 d

300G

DH

act

ivity

(U/I)

200

100

Portal veinHepatic veinHeart

*

** *

00 h 1 h 6 h 12 h 1 d 2 d 3 d 4 d 6 d 12 d

45

α-ke

togl

utar

ate

(μM

)

30

15

**

*

**

*

*

*

* * * *

00 h 1 h 6 h 12 h 1 d 2 d 3 d 4 d 6 d 12 d

200

Glu

tam

ate

(μM

) 150

100

50

Time after CCl4 administration

*

*

A B

0

150

200

*

**

Amm

onia

(μM

)

50

100

0

600 *

**

*

Glu

tam

ate

(μM

)

200

400

0

6000***

GD

H a

ctiv

ity (U

/I)

2000

4000

α-KG - + + + + +AOA - - + + + +

NADPH - - - + + +GDH - - - - + +

PDAC - - - - - +

0

*

Amm

onia

(μM

)

1000

400

200

600

800

0

**

Glu

tam

ate

(μM

)

500

200

100

300

400

0

GD

H a

ctiv

ity (U

/I)

300

200

100

***

ControlNH4ClNH4Cl + α-KGNH4Cl + α-KG+ PDAC

C

D Normal situation

Liver

Blood

PericentralPeriportalGlnase

Gln

Gln

GS

GDHGDH

Urea

Ureacycle

Glu Glu

NH4+

NH4+

NH4+

NH4+

α-KGin α-KGout

α-KG

α-KG

α-KG+

α-KX

*

ductio isutama erever l4PH, GD As adde Gked by eConc alperice eriporta ale, the isoweve dan go e

Research Article

JHEPAT 5901 No. of Pages 13

8 January 2016

Brain

(Low; <50 μM)

No toxicity

Fig. 3. Detoxification of ammonia by a reverse GDH reaction. (A) After inaccompanied by a decrease in alpha-ketoglutarate (a-KG) and an increase in globserved in liver tissue (Supplementary Tables 1 and 2). (B) Validation of theinjection was analyzed. a-KG was added alone or in combination with AOA, NADsimilar experimental design was chosen as in B. However, 600 lM ammonia wadecreases ammonia and increases glutamate concentrations, which can be blocNH4Cl group (0 h). Data are mean values and SD of 3 biological replicas. (D)hepatocytes, GDH generates ammonia, which is detoxified by the urea cycle. InGS reaction to form glutamine (Gln). Biosynthesis of a-KG takes place in the pehepatocytes, where it is needed for GS [25,26]. After induction of liver damagcompletely destroyed. This leads to increased blood ammonia concentrations. Hconsuming ammonia and a-KG to generate glutamate (Glu). This reaction ctherapeutically substituted.

Please cite this article in press as: Ghallab A et al. Model-guided identificatioHepatol (2016), http://dx.doi.org/10.1016/j.jhep.2015.11.018

6 Journal of Hepatology 2

After liver damage: depletion of α-KG

(Hyperammonemia)

(depletion)

PericentralPeriportalGlnase

(depletion)(compromised)

Gln

Gln

GS

GDH

GDH GDH

GDH

Urea

Urea cycle

Glu Glu

NH4+

NH4+

NH4+

NH4+

Gin α-KGout

α-KG

α-KG+

Risk of hepaticencephalopathy

X

X

X

GDHGluα-KG

n of liver damage by CCl4, plasma activity of GDH transiently increases. Thiste. ⁄p <0.05 when compared to the corresponding control (0 h). Similar results werse GDH reaction using an inhibitor of GDH (PDAC). Plasma of mice 24 h after CCH and PDAC. ⁄⁄⁄p <0.001, ⁄⁄p <0.01 and ⁄p <0.05 when compared to controls (0). (C)d to also test the reaction at a higher but still clinically relevant concentration. a-Kthe GDH inhibitor, PDAC. ⁄⁄⁄p <0.001, ⁄⁄p <0.01 and ⁄p <0.05 when compared to th

ept of the reverse glutamate dehydrogenase (GDH) reaction. In normal periportntral hepatocytes, GDH generates glutamate which is required as a substrate for thl hepatocytes; a-KG is then partially exported and taken up again by the pericentrexpression of urea cycle enzymes decreased and the pericentral region with GSr, also GDH is released from damaged hepatocytes and catalyzes a reaction in bloountil a-KG in blood is consumed. In this situation a-KG and NADPH should b

n of a therapeutic strategy to reduce hyperammonemia in liver diseases. J

016 vol. xxx j xxx–xxx

296

297 ta298 hi299 re300 bu301 th302 in303 tio304 am305 tio306 m307

308 ev309 ho310 N

311centrations of both NADPH and NADP+ increased after induction312of liver damage by CCl4 (Supplementary Fig. 15A). In addition, an313enhanced NADP+/NADPH ratio was observed in both blood and314liver tissue (Supplementary Fig. 15B). This increase in NADP+/315NADPH ratio fits to a switch in the GDH reaction from NADPH316generation to NADPH consumption. Despite the increase in317N318st319pl320ba321N322l)323am324in325fo326ce327AO328pl329G330tr331th332tr333ic334re

A

* ***

0

15

25

*****

**

GD

H a

ctiv

ity (U

/g p

rote

in)

5

20

10

Control CCl4 (1.6 g/kg)-PDAC+PDAC

-PDAC+PDAC

**ºº

**

º

*0

1500

2500

*

Amm

onia

(μM

)

500

2000

1000

B

C

00.0

60

***

Glu

tam

ate

(μM

)

20

40

0.5 2.0NH4Cl (mM)

**

*

º

º

º

0.0 0.5 2.0NH4Cl (mM)

D

0

80

**

Glu

tam

ate

(μM

)

20

40

60ControlCCl4

E

0

250

*

Ure

a (μ

M)

50

100

150

200 ControlCCl4

Fig. 4. Ammonia consumption by GDH in primary mouse hepatocytes.HemdiComto(Dinm

A

Blood far periportal

Cellular farperiportal

Cellularmid-periportal

Cellular pericentral

GLNase GDH

CPS

Gln

GlnUrea

Glu

α-KGNH4

Blood mid-periportal Blood pericentral

GDH

GDH

CPS GS

Urea

Glu

Gluα-KG

α-KG

NH4NH4

00 h 1 h 6 h 12 h 1 d 2 d 3 d 4 d 6 d 12 d

500

Amm

onia

(μM

)

200

300

100

400

meas, in

Time after CCl4 administration

meas, outsim, out

B

Fig. 5. Integration of the GDH reaction into the metabolic model. (A)Scheme of the metabolic reactions and zones of the extended model includingGDH in the blood of the liver and hepatocytes. (B) The model extension leads to abetter fit between simulated and experimental data.

JOURNAL OF HEPATOLOGY

PH

JHEPAT 5901 No. of Pages 13

8 January 2016

In the aforementioned experiment, the molar amount of glu-mate produced in the damaged liver after a-KG injection wasgher than ammonia consumption (Fig. 6A). Therefore, thesults cannot only be explained by the reverse GDH reaction,t may be due to the consumption of a-KG by transaminasesat contribute to the generation of glutamate. Indeed, tail veinjection of the transaminases inhibitor AOA prior to a-KG injec-n reduced the production of glutamate (Fig. 6C) and improved

patocytes were isolated from CCl4 (1.6 g/kg) intoxicated (day 1) and untreatedice and suspended at a concentration of 2 million hepatocytes/ml for 1 h withfferent concentrations of ammonia. (A) Inhibition of GDH activity by PDAC. (B)mpromised ammonia detoxification after GDH inhibition. (C) Reduced gluta-ate production by GDH inhibition. ⁄⁄⁄p <0.001, ⁄⁄p <0.01 and ⁄p <0.05 compared– PDAC. ��p <0.01 and �p <0.05 compared to hepatocytes from untreated mice.& E) compromised urea and glutamine production by hepatocytes of CCl4

toxicated mice. ⁄⁄p <0.01 and ⁄p <0.05 compared to hepatocytes from untreatedice. Data are mean values and SD of three independent experiments.

monia detoxification. The efficiency of transaminases inhibi-n by AOA in vivo has been confirmed in preliminary experi-

335

336an337in338in339ad340(2341ve

ents (Supplementary Figs. 13 and 14).The reverse GDH reaction requires NADPH as a cofactor; how-er, NADPH concentrations are very low in blood. To determinew NADPH levels are altered in our model of liver damage, bothADPH and its oxidized form NADP+ were analyzed. Blood con-

lease cite this article in press as: Ghallab A et al. Model-guided identification ofepatol (2016), http://dx.doi.org/10.1016/j.jhep.2015.11.018

Journal of Hepatology 2016

ADPH after induction of liver damage, the concentrations areill relatively low. Therefore, to study the influence of NADPH,asma from mice collected 24 h after CCl4 injection was incu-ted with varying concentrations of NADPH in the presence ofH4Cl (1 mM), a-KG (3 mM), AOA (1 mM) and GDH (12,000 U/for one hour. A concentration dependent decrease in plasmamonia and an increase in glutamate were observed with

creasing concentrations of NADPH (Fig. 7A). A similar trendr ammonia and glutamate was observed with increasing con-ntrations of a-KG and GDH (Fig. 7B, C). Moreover, addition ofA reduced both ammonia and glutamate concentrations (Sup-ementary Fig. 16). To understand how the orientation of theDH reaction is controlled by ammonia and glutamate concen-ations, titration experiments were performed, which indicatedat GDH significantly consumes ammonia beginning at concen-ations of 150 lM and higher (Fig. 7D). In contrast, unphysiolog-ally high concentrations of more than 10 mM glutamate werequired to block the reaction (Fig. 7E).Based on these in vitro optimized concentrations, we designedin vivo study to treat hyperammonemia in mice. After the

duction of liver damage by CCl4, transaminases activities werehibited by AOA (13 mg/kg; tail vein injection; 24 h after CCl4ministration). Thirty minutes later a cocktail of a-KG80 mg/kg), GDH (720 U/kg) and NADPH (180 mg/kg) was intra-nously injected. A dose of 280 mg/kg a-KG was chosen because

a therapeutic strategy to reduce hyperammonemia in liver diseases. J

vol. xxx j xxx–xxx 7

342 it transiently normalized a-KG levels in mice 24 h after CCl4.343 720 U/kg GDH was used because it resulted in plasma levels of344 approximately 6000 U/l 15 min after injection (Supplementary345 Fig. 17), an activity level shown to allow maximal ammonia con-346 sumption in plasma in vitro ( Fig. 7C). The dose of 180 mg/kg347 NADPH was also considered as adequate in a pharmacokinetic348 experiment (Supplementary Fig. 18) as it transiently increased349 plasma NADPH to approximately 1.6 mM 2 min after injection.350 Injection of the a-KG/GDH/NADPH cocktail (KGN cocktail)351 reduced ammonia concentrations from 213 to 74 lM within352 15 min after administration (Fig. 8). Simultaneously, glutamate353 levels increased from 131 to 369 lM. Analysis of a-KG and354 GDH activity in the plasma showed that substitution was suc-355 cessful 15 min after the injection of the KGN cocktail. Moreover,356 the activities of aspartate and alanine aminotransferase were suc-357 e358 s.

359

360 a361 e362

363can be used therapeutically by the administration of a cocktail364of GDH and cofactors.365Therapy for hyperammonemia remains challenging [8–10].366Hemodialysis is the most efficient treatment for reducing ele-367vated blood ammonia concentrations [11,12]. For milder forms368of hyperammonemia, pharmacologic management is possible369[13]. Efficient strategies for patients with urea cycle defects370include infusion of phenylacetate or benzoate. Phenylacetate371combines with glutamine to form a product which can be372excreted by the kidneys [9,13,14]. Conversely, benzoate combines373with glycine to form hippurate, which is also excreted in urine374[13,15]. Both compounds reduce the total body nitrogen content;375however, this therapy has also failed in a fraction of patients with376hyperammonemic crisis who became refractory most probably377due to the accumulation of nitrogen waste [9]. This led to the378

379e380l381-382e383n384

385-386s.

AC

Cl 4 t

reat

ed m

ice

Time after α-KG injection (min)

00 15 30 60

150

450

***Am

mon

ia (μ

M)

300

00 15 30 60

150

450

*

***

Glu

tam

ate

(μM

)

300

00 15 30 60

20

80

α-Ke

togl

utar

ate

(μM

)

40

60

00 15 30 60

100

400

GD

H a

ctiv

ity (U

/I)

200

300

B

Con

trol m

ice

00 15 30 60

30

90

Amm

onia

(μM

)

600

00 15 30 60

20

80

Glu

tam

ate

(μM

)40

60

00 15 30 60

300

1200 ***

*

α-Ke

togl

utar

ate

(μM

)

600

900

00 15 30 60

20

40

GD

H a

ctiv

ity (U

/L)

Time after α-KG infusion (min)

C

CC

l 4 a

nd A

OA

treat

ed m

ice

Time after α-KG injection (min)

0

300

Amm

onia

(μM

)

100

200

-30 0 150

800

*

Glu

tam

ate

( μM

)

200

400

600

-30 0 15

**

0

1000

α-Ke

togl

utar

ate

(μM

)

200

400

600

800

-30 0 150

Glu

tam

ate

(μM

)

100

200

300

-AOA+AOA

-30 0 15

Fig. 6. Reduction of blood ammonia concentrations by a-KG. (A) Tail vein injection of 280 mg/kg a-KG into mice 24 h after induction of liver damage by CCl4 (1.6 g/kg).(B) Control experiment with a-KG (280 mg/kg) injected into the tail vein of untreated mice. ⁄⁄⁄p <0.001, ⁄⁄p <0.01 and ⁄p <0.05 when compared to the control group (0). (C)Influence of the transaminase inhibitor, AOA (13 mg/kg; tail vein injection) on ammonia detoxification by a-KG. ⁄⁄p <0.01 and ⁄p <0.05 when compared to thecorresponding control. Data are mean values and SD of three mice treated at different experimental days with individually prepared a-KG.

Research Article

JHEPAT 5901 No. of Pages 13

8 January 2016

cessfully inhibited by AOA. The mice were observed for threweeks after the experiment and did not show any complication

Discussion

Guided by simulations with an IM predicting a missing ammonisink after severe CCl4-induced liver damage, we identified th

GDH reaction as fundamental for ammonia consumption, which

Please cite this article in press as: Ghallab A et al. Model-guided identificatioHepatol (2016), http://dx.doi.org/10.1016/j.jhep.2015.11.018

8 Journal of Hepatology 2

concept that only blood ammonia concentrations below 500 lMshould be treated pharmacologically; whereas, more severhyperammonemia requires aggressive interventions with renareplacement therapies, such as hemodialysis [12,16]. In such situations with either severe or refractory hyperammonemia ththerapeutic strategy developed in the present study may be aalternative to hemodialysis.

The current experiments demonstrate that infusion of a KGNcocktail reduces ammonia close to normal levels within minute

n of a therapeutic strategy to reduce hyperammonemia in liver diseases. J

016 vol. xxx j xxx–xxx

387 U388 bl389 m390 re391 co392 G393 In394 am

395Th396re397consuming reaction under conditions of hyperammonemia. By398re399da400ni401av402th

A

0

300

1500

**

******

*** ***Amm

onia

(μM

)

900

600

1200

0- + + +

300

1500

******

*** *** ***

Glu

tam

ate

(μM

)

+ + + +

900

600

1200

NH 4Cl (1

mM)

- - + + + + + +

Cofacto

rs- - -

125

250

500

1000

2000

NADPH (μM)

B

0

500

1500

*****

Amm

onia

(μM

)

1000

0- + + +

500 *

Glu

tam

ate

(μM

)

+

1500

1000

2000

- - + + +- - -

375

75NH 4Cl (1

mM)

Cofacto

rs

α-KG (μ

M)

D

0

750

Glu

tam

ate

( μM

)

150

450

300

600

NH4Cl (μM)37.5 75 150 300 600

***

***-Cocktail+Cocktail

0

750

Amm

onia

(μM

)

150

450

300

600-Cocktail+Cocktail

* ** ***

37.5 75 150 300 600

E

Fi ationin 4Cl (1GD cted fco M), Naft of NHfo plasmDi tail ofan ompared – cocktail. (E) Ammonia (600 lM) detoxification by the cocktail was onlybl o NH4Cl alone. Data are mean values and SD of three biological replicas.

JOURNAL OF HEPATOLOGY

PH

JHEPAT 5901 No. of Pages 13

8 January 2016

nder these conditions, a GDH-catalyzed reaction takes place inood where ammonia and a-KG are consumed to form gluta-ate in an NADPH-dependent reaction. GDH was also previouslyported to switch its catalytic orientation under physiologicalnditions. In the periportal compartment of the liver lobule,DH generates ammonia (Fig. 3D), which fuels the urea cycle.the pericentral compartment, GDH is known to consumemonia to generate glutamate for the GS reaction [17–19].

g. 7. Optimization of cofactor concentrations for the GDH reaction and identificjection was incubated with varying concentrations of NADPH in the presence of NHH (12,000 U/L) for one hour. (B) Alpha-ketoglutarate (a-KG): plasma was collencentrations of a-KG in the presence of NH4Cl (1 mM) and other cofactors; AOA (1 mer CCl4 injection was incubated with varying concentrations of GDH in the presencer one hour. ⁄⁄⁄p <0.001, ⁄⁄p <0.01 and ⁄p <0.05 compared to the situation where thefferent ammonia concentrations were added to plasma of untreated mice and a cockalyze ammonia and glutamate one hour later. ⁄⁄⁄p <0.001, ⁄⁄p <0.01 and ⁄p <0.05 cocked by unphysiologically high glutamate concentrations. ⁄⁄⁄p <0.001 compared t

lease cite this article in press as: Ghallab A et al. Model-guided identification ofepatol (2016), http://dx.doi.org/10.1016/j.jhep.2015.11.018

Journal of Hepatology 2016

*

******

**

***

+ ++ +

015

0030

00

C

0

300

1500

* **

******

*** ***Amm

onia

(μM

)

900

600

1200

0- + + +

300***

**

***

******

*** ***

Glu

tam

ate

(μM

)

+ + +

900

1200

600

1500

- - + + + + +- - -

188

375

750

1500

++

3000

++

6000

++

12,00

0NH 4Cl (1

mM)

Cofacto

rs

GDH (U/L)

0

200

400

800

*** ** *****Am

mon

ia (μ

M)

600

0

10

20

40

Glu

tam

ate

(mM

)

30

+ + + + + +- + + + + +- -

3.16

10.00

31.60

NH4ClCocktail

Glutamate (mM)1.0

0

of maximal GDH activity. (A) NADPH: mouse plasma collected 24 h after CCl4mM) and other cofactors; aminooxy acetate (AOA) (1 mM), a-KG (3 mM) and

rom mice on day one after CCl4 administration and incubated with varyingADPH (1 mM) and GDH (12,000 U/L). (C) GDH: plasma frommice collected 24 h4Cl (1 mM) and other cofactors; AOA (1 mM), a-KG (3 mM), and NADPH (1 mM)a was incubated with only NH4Cl. (D) Ammonia detoxification by the cocktail.a-KG (3 mM), NADPH (0.5 mM), GDH (6000 U/L) and AOA (1 mM) was added to

e present study shows by use of a GDH inhibitor that GDHleased from damaged liver tissue may catalyze an ammonia

leasing GDH from damaged hepatocytes into the blood, themaged liver provides a mechanism that reduces blood ammo-a levels. However, this protective mechanism is limited by theailability of the GDH substrate, a-KG. The present study showsat a-KG strongly decreases upon induction of acute liver dam-

a therapeutic strategy to reduce hyperammonemia in liver diseases. J

vol. xxx j xxx–xxx 9

403 s404 s405 -406 -407 -408 r409 -410 n411 e412 -413 -414 4

415 d416 l-417 l418 .419 s

420

421

422a-KG was previously tested for the treatment of hyperam-423monemia between 1964 and 1978 [21,22], but this strategy424i-425s426n427i-428y429a relatively weak reduction of hyperammonemia. This reduction430may be possible because in addition to GDH, NADPH is also431

432/433d434n435s436d437.438n439n440g441n442n443-444d445y446S4474,

448s449a450-451-452g453e454d455s.456s457e458o459-460n461-462d463

464t465n466l-467a468

469

470l471

472

473y474f475

ddril,ofndDde

Q5 Q6

Research Article

JHEPAT 5901 No. of Pages 13

8 January 2016

age, because it is consumed by the reverse GDH reaction. Thiprompted us to supplement a-KG, increase blood concentrationof NADPH, and infuse GDH. Indeed, the combined injection of aKG, GDH and NADPH efficiently reduced blood ammonia concentrations. A GDH bolus was injected to result in plasma peak concentrations between 5000 and 6000 U/L. This is in the same ordeof magnitude as observed in patients after acetaminophen intoxication [20]. Therefore, in patients with acute liver intoxicatiowith high blood GDH, therapy with a-KG and NADPH might bsufficient. However, it should be considered that the GDH reaction in blood described here (Fig. 3D) does not explain all experimental observations: a-KG decreases significantly 12 h after CCladministration when there is no significant increase in blooGDH (Fig. 3A). This discrepancy may be explained by the intracelular change of the catalytic direction of GDH in the periportahepatocytes, which may precede the GDH release into the bloodHowever, this was not further analyzed in the present study a

0

50

100

520Am

mon

ia (μ

M)

150

200

***

0

200

400

600

Glu

tam

ate

(μM

)

***

- - +- + +

KGN-cocktailAOA

0

300

600

900

1200

α-Ke

togl

utar

ate

( μM

)

Time after KGNinjection (min)

*

-30 0 15

0

2000

4000

8000

AST

(U/L

)

6000

******

0

4000

8000

16000

12000AL

T (U

/L)

**

- - +- + +

KGN-cocktailAOA

0

2000

4000

6000

GD

H a

ctiv

ity (U

/L)

Time after KGNinjection (min)

***

-30 0 15

A

Fig. 8. Treatment of hyperammonemia by injection of a cocktail of GDH anoptimized cofactor doses. A cocktail of GDH (720 U/kg), a-KG (280 mg/kg) anNADPH (180 mg/kg) (KGN) was injected into mice 24 h after induction of livedamage using CCl4 (1.6 g/kg). Thirty minutes prior to treatment with the cocktamice received a single dose of 13 mg/kg AOA to block transaminases. Injectionthe KGN cocktail reduced ammonia and increased glutamate concentrations ithe blood of mice. a-KG and GDH activity increased while aspartate (AST) analanine aminotransferase (ALT) activities decreased. Data are mean values and Sof four mice treated at different experimental days with individually preparetherapeutic cocktails. ⁄⁄⁄p <0.001, ⁄⁄p <0.01 and ⁄p <0.05 when compared to thcontrol group (�30).

Please cite this article in press as: Ghallab A et al. Model-guided identificatioHepatol (2016), http://dx.doi.org/10.1016/j.jhep.2015.11.018

10 Journal of Hepatology 2

we choose to focus on the therapeutically more relevant GDHreaction taking place in blood.

was abandoned because it was not sufficiently efficient for clincal application. The reverse GDH reaction in the blood and itrequirement for NADPH as a cofactor was not yet known whethe early therapeutic studies with a-KG were performed. In addtion, injection of a-KG alone in the present study resulted in onl

released from deteriorating hepatocytes after CCl4 injection.The concept of treating hyperammonemia by a-KG/GDH

NADPH infusion originates from simulations using an integratemetabolic spatio-temporal model [4]. This model is based owell-understood pathways of ammonia detoxification, such aurea cycle enzymes and the GS reaction [23,24]. It predictehigher ammonia concentrations compared to the measured dataTherefore, we analyzed liver tissue during the damage inductioand regeneration processes, but the results could also not explaithe discrepancy. Time-resolved gene array experiments followinCCl4 injection led to the observation that a general decrease imetabolizing enzymes occurs including enzymes involved iammonia metabolism. All enzymes of the urea cycle were transcriptionally downregulated by at least 60%. Factors identifieby the gene array analysis were further analyzed by activitassays and immunostaining. Key observations were: (a) the Gpositive region, which is initially completely destroyed by CClshows a delayed recovery and does not return to normal levelbefore day 12; and (b) CPS1, the rate limiting enzyme in the urecycle normally expressed in the periportal region, is downregulated during the destruction process (days 1–3), but its expression then extends throughout the entire liver lobule durindays 4–6. The other urea cycle enzymes showed a similar timcourse as CPS1 with the exception of arginase1, which decreaseonly slightly during the destruction and regeneration procesGlutaminase showed a similar time course and pattern aCPS1. Nevertheless, none of these alterations could explain thobserved discrepancy. However, the refined models that take intaccount the reversible GDH reaction show an excellent agreement with the experimental data suggesting that consumptioby the GDH reaction represents the previously predicted ammonia sink, hence providing an example for model guideexperimentation.

In conclusion, a novel form of therapy has been identified thaallows the rapid correction of hyperammonemia by the infusioof a-KG, GDH and NADPH. This pharmacotherapy may prove reevant as an emergency therapy for episodes of hyperammonemiin urea cycle disease or liver cirrhosis.

Financial support

This study was supported by the BMBF funded projects VirtuaLiver (FKZ0315739), Lebersimulator and FP 7EU NOTOX.

Conflict of interest

The authors who have taken part in this study declared that thedo not have anything to disclose regarding funding or conflict ointerest with respect to this manuscript.

n of a therapeutic strategy to reduce hyperammonemia in liver diseases. J

016 vol. xxx j xxx–xxx

476 Au

477478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529 Ac

530 W531 Br532 En533 D534 Ta

535fo536Co

537Su

538Su539th54001

541References

JOURNAL OF HEPATOLOGY

PH

JHEPAT 5901 No. of Pages 13

8 January 2016

thors’ contributions

Ahmed Ghallab: study concept and design; acquisition ofdata; analysis and interpretation of data; drafting of themanuscript; statistical analysis; critical revision of themanuscript.Sebastian G. Henkel: mathematical modeling (integratedmodel of Schliess et al. (2014), model 0 and extrahepatic massbalance); analysis and interpretation of data; drafting of themanuscriptGéraldine Cellière: mathematical modeling (novel ammoniadetoxification models, extension of model 0); analysis andinterpretation of data; drafting of the manuscript, statistical

analysis.Dominik Driesch: mathematical modeling (integrated modelof Schliess et al. (2014), model 0 and extrahepatic mass bal-ance); analysis and interpretation of data; drafting of themanuscript.Stefan Hoehme: mathematical modeling (integrated model ofSchliess et al. (2014), model 0); analysis and interpretation ofdata; drafting of the manuscript.Ute Hofmann: acquisition of data; analysis and interpretationof data; technical support; drafting of the manuscript.Sebastian Zellmer: study concept and design; acquisition ofdata; analysis and interpretation of data; drafting of themanuscript; critical revision of the manuscript.Patricio Godoy: acquisition of data; analysis and interpreta-tion of data; drafting of the manuscript.Agapios Sachinidis: acquisition of data; analysis and interpre-tation of data; drafting of the manuscript.Meinolf Blaszkewicz: acquisition of data; analysis and inter-pretation of data; technical support; drafting of themanuscript.Raymond Reif: analysis and interpretation of data; drafting ofthe manuscript.Rosemarie Marchan: critical revision of the manuscript.Lars Kuepfer: mathematical modeling; drafting of themanuscript.Dieter Häussinger: study concept and design; critical revisionof the manuscript.Dirk Drasdo: study concept and design; mathematical model-ing (integrated model of Schliess et al. (2014), model 0 andnovel models); analysis and interpretation of data; draftingof the manuscript, statistical analysis, critical revision of themanuscriptRolf Gebhardt: study concept and design; acquisition of data;analysis and interpretation of data; drafting of the manuscript,critical revision of the manuscript.Jan G. Hengstler: study concept and design; acquisition ofdata; analysis and interpretation of data; drafting of themanuscript, statistical analysis, critical revision of the manu-script; study supervision.

knowledgments

e thank Mrs. Gabi Baumhoer, Mrs. Georgia Günther and Mrs.igitte Begher-Tibbe – Leibniz Research Centre for Workingvironment and Human Factors at the Technical Universityortmund, Dortmund, Germany; and Ms. Margit Henry and Ms.mara Rotshteyn – Institute of Neurophysiology and Center

Au

[

[

[

[

[

[

[

[

[

[1

[1

[1

[1

[1

[1

[1

[1

lease cite this article in press as: Ghallab A et al. Model-guided identification ofepatol (2016), http://dx.doi.org/10.1016/j.jhep.2015.11.018

Journal of Hepatology 2016

r Molecular Medicine Cologne (CMMC), University of Cologne,logne, Germany – for competent technical assistance.

pplementary data

pplementary data associated with this article can be found, ine online version, at http://dx.doi.org/10.1016/j.jhep.2015.11.8.

542thor names in bold designate shared co-first authorship

5431] Drasdo D, Hoehme S, Hengstler JG. How predictive quantitative modelling of544tissue organisation can inform liver disease pathogenesis. J Hepatol5452014;61:951–956.5462] Zellmer S, Schmidt-Heck W, Godoy P, Weng H, Meyer C, Lehmann T, et al.547Transcription factors ETF, E2F, and SP-1 are involved in cytokine-548independent proliferation of murine hepatocytes. Hepatology 2010;52:5492127–2136.5503] Godoy P, Hewitt NJ, Albrecht U, Andersen ME, Ansari N, Bhattacharya S, et al.551Recent advances in 2D and 3D in vitro systems using primary hepatocytes,552alternative hepatocyte sources and non-parenchymal liver cells and their553use in investigating mechanisms of hepatotoxicity, cell signaling and ADME.554Arch Toxicol 2013;87:1315–1530.5554] Schliess F, Hoehme S, Henkel SG, Ghallab A, Driesch D, Bottger J, et al.556Integrated metabolic spatial-temporal model for the prediction of ammonia557detoxification during liver damage and regeneration. Hepatology5582014;60:2040–2051.5595] Hoehme S, Brulport M, Bauer A, Bedawy E, Schormann W, Hermes M, et al.560Prediction and validation of cell alignment along microvessels as order561principle to restore tissue architecture in liver regeneration. Proc Natl Acad562Sci U S A 2010;107:10371–10376.5636] Campos G, Schmidt-Heck W, Ghallab A, Rochlitz K, Putter L, Medinas DB,564et al. The transcription factor CHOP, a central component of the565transcriptional regulatory network induced upon CCl4 intoxication in566mouse liver, is not a critical mediator of hepatotoxicity. Arch Toxicol5672014;88:1267–1280.5687] Gebhardt R, Baldysiak-Figiel A, Krugel V, Ueberham E, Gaunitz F. Hepato-569cellular expression of glutamine synthetase: an indicator of morphogen570actions as master regulators of zonation in adult liver. Prog Histochem571Cytochem 2007;41:201–266.5728] Levesque R, Leblanc M, Cardinal J, Teitlebaum J, Skrobik Y, Lebrun M.573Haemodialysis for severe hyperammonaemic coma complicating urinary574diversions. Nephrol Dial Transplant 1999;14:458–461.5759] Enns GM, Berry SA, Berry GT, Rhead WJ, Brusilow SW, Hamosh A. Survival576after treatment with phenylacetate and benzoate for urea-cycle disorders. N577Engl J Med 2007;356:2282–2292.5780] Poh Z, Chang PE. A current review of the diagnostic and treatment strategies579of hepatic encephalopathy. Int J Hepatol 2012;2012 480309.5801] Clay AS, Hainline BE. Hyperammonemia in the ICU. Chest5812007;132:1368–1378.5822] Rajpoot DK, Gargus JJ. Acute hemodialysis for hyperammonemia in small583neonates. Pediatr Nephrol 2004;19:390–395.5843] Summar M. Current strategies for the management of neonatal urea cycle585disorders. J Pediatrics 2001;138:S30–S39.5864] Honda S, Yamamoto K, Sekizuka M, Oshima Y, Nagai K, Hashimoto G, et al.587Successful treatment of severe hyperammonemia using sodium phenylac-588etate powder prepared in hospital pharmacy. Biol Pharm Bull5892002;25:1244–1246.5905] Misel ML, Gish RG, Patton H, Mendler M. Sodium benzoate for treatment of591hepatic encephalopathy. Gastroenterol Hepatol 2013;9:219–227.5926] Collen JF, Das NP, Koff JM, Neff RT, Abbott KC. Hemodialysis for hyperam-593monemia associated with ornithine transcarbamylase deficiency. Appl Clin594Genet 2008;1:1–5.5957] Boon L, Geerts WJ, Jonker A, Lamers WH, Van Noorden CJ. High protein diet596induces pericentral glutamate dehydrogenase and ornithine aminotrans-597ferase to provide sufficient glutamate for pericentral detoxification of598ammonia in rat liver lobules. Histochem Cell Biol 1999;111:445–452.

a therapeutic strategy to reduce hyperammonemia in liver diseases. J

vol. xxx j xxx–xxx 11

599 [18] Sies H, Häussinger D, Grosskopf M. Mitochondrial nicotinamide nucleotide600 systems: ammonium chloride responses and associated metabolic transi-601 tions in hemoglobin-free perfused rat liver. Hoppe Seylers Z Physiol Chem602 1974;355:305–320.603 [19] Spanaki C, Plaitakis A. The role of glutamate dehydrogenase in mammalian604 ammonia metabolism. Neurotox Res 2012;21:117–127.605 [20] McGill MR, Sharpe MR, Williams CD, Taha M, Curry SC, Jaeschke H. The606 mechanism underlying acetaminophen-induced hepatotoxicity in humans607 and mice involves mitochondrial damage and nuclear DNA fragmentation. J608 Clin Invest 2012;122:1574–1583.609 [21] Meuret G, Beck K, Keul J, Gruenagel HH. On therapy of hepatic coma with610 metabolites of the urea cycle, glutamate and alpha-ketoglutarate. Dtsch Med611 Wochenschr 1968;93:1194–1197.612 [22] Wildhirt E. The terminal stage of liver diseases and hepatic coma. Der613 Internist 1965;6:439–446.

614[23] Gebhardt R, Mecke D. Heterogeneous distribution of glutamine synthetase615among rat liver parenchymal cells in situ and in primary culture. EMBO J6161983;2:567–570.617[24] Häussinger D. Hepatocyte heterogeneity in glutamine and ammonia618metabolism and the role of an intercellular glutamine cycle during619ureogenesis in perfused rat liver. Eur J Biochem 1983;133:269–275.620[25] Stoll B, Häussinger D. Hepatocyte heterogeneity in uptake and metabolism621of malate and related dicarboxylates in perfused rat liver. Eur J Biochem6221991;195:121–129.623[26] Stoll B, McNelly S, Buscher HP, Häussinger D. Functional hepatocyte624heterogeneity in glutamate, aspartate and alpha-ketoglutarate uptake: a625histoautoradiographical study. Hepatology 1991;13:247–253.

626

Research Article

Please cite this article in press as: Ghallab A et al. Model-guided identification of a therapeutic strategy to reduce hyperammonemia in liver diseases. JHepatol (2016), http://dx.doi.org/10.1016/j.jhep.2015.11.018

12 Journal of Hepatology 2016 vol. xxx j xxx–xxx

JHEPAT 5901 No. of Pages 13

8 January 2016