Derangements of liver tissue bioenergetics in Concanavalin A-induced hepatitis.

1.8 MB·English

by  Al-Shamsi, Mariam

Al-Shamsietal.BMCGastroenterology2013,13:6
http://www.biomedcentral.com/1471-230X/13/6
RESEARCH ARTICLE Open Access
Derangements of liver tissue bioenergetics in
Concanavalin A-induced hepatitis
Mariam Al-Shamsi1, Allen Shahin1, Eric PK Mensah-Brown2* and Abdul-Kader Souid3*
Abstract
Background: Anovelin vitro system was employed to investigate liver tissue respiration (mitochondrial O
2
consumption) inmice treated with concanavalin A (Con A). This study aimed to investigate hepatocyte
bioenergetics in this well-studied hepatitis model.
Methods: C57Bl/6and C57Bl/6 IFN-γ−/− mice were injected intravenously with 12 mg ConA/kg. Liver specimens
were collectedatvarious timepoints after injection and analyzedfor cellular respirationand caspase activation.
Serum was analyzedfor interferon-gamma (IFN-γ) and aminotransferases. Fluorescenceactivated cell sorting
analysis was used to determine the phenotype of infiltrating cells, and light and electron microscopy were used to
monitor morphological changes.Phosphorescenceanalyzerthat measured dissolved O as function oftime was
2
used to evaluate respiration.
Results: Insealed vials, O concentrations in solutions containing liver specimen and glucose declinedlinearly with
2
time, confirming zero-order kinetics of hepatocyte respiration. O consumptionwas inhibitedby cyanide,
2
confirming the oxidation occurred inthe respiratory chain. Enhanced liver respiration (by≈68%, p<0.02) was noted
3hr after ConA treatment, and occurred inconjunctionwith limited cellularinfiltrations around theblood vessels.
Diminished respiration (by≈30%, p=0.005) was noted 12 hr after ConAtreatment, and occurred inconjunction
withderangedmitochondria,areasofnecrosis,andprominentinfiltrationswithimmunecells,mostsignificantly,
CD3+NKT+cells.IncreasesinintracellularcaspaseactivityandserumIFN-γandaminotransferaselevelswerenoted
3hrafterConAtreatmentandprogressedwithtime.Theabove-notedchangeswerelesspronouncedinC57Bl/6
IFN-γ−/−micetreatedwithConA.
Conclusions:Basedontheseresults,livertissuebioenergeticsisincreased3hrafterConAexposure.Thiseffectis
drivenbythepathogenesisofthedisease,inwhichIFN-γandothercytokinescontributeto.Subsequentdeclines
inliverbioenergeticsappeartobearesultofnecrosisandactivecaspasestargetingthemitochondriawithin
hepatocytes.
Keywords:ConcanavalinA(ConA),Hepatitis,Caspase,AST,ALT,IFN-γ
Background oxidation of reduced metabolic fuels with the passage of
Theterm“cellularbioenergetics”describesthebiochemical electrons to O , and the synthesis of ATP. Impaired cellu-
2
processes involved in energy metabolism (energy conver- lar bioenergetics or respiration thus entails an interference
sionortransformation),andtheterm“cellularrespiration” withanyofthesemetabolicprocesses.
(mitochondrialoxygenconsumption)describesthedelivery Concanavalin A (ConA) is a plant lectin from seeds of
of metabolites and oxygen (O ) to the mitochondria, the Canavalia ensiformis [Jack bean] that serves as a poly-
2
clonal T-cell mitogen and is also used to induce hepatitis
in mice. Murine ConA-induced hepatitis is a well-studied
model that mimics human liver viral infections [1,2]. The
*Correspondence:[email protected];[email protected]
2DepartmentofAnatomy,CollegeofMedicineandHealthSciences,United liver injury is typically noted within 3 hr of intravenous
ArabEmiratesUniversity,AlAin,AbuDhabi,UnitedArabEmirates(UAE) injection of ConA (12 mg/kg) and progresses with time
3DepartmentofPediatrics,CollegeofMedicineandHealthSciences,United
[2]. Activation and recruitmentof natural killer T (NK T)
ArabEmiratesUniversity,AlAin,AbuDhabi,UnitedArabEmirates(UAE)
Fulllistofauthorinformationisavailableattheendofthearticle
©2013Al-Shamsietal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative
CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and
reproductioninanymedium,providedtheoriginalworkisproperlycited.
Al-Shamsietal.BMCGastroenterology2013,13:6 Page2of13
http://www.biomedcentral.com/1471-230X/13/6
and other cells of the innate immune system are early the principle that O quenches the phosphorescence of
2
events that, in turn, lead to secretion of inflammatory palladium (Pd) II-meso-tetra-(4-sulfonatophenyl)-tetra-
cytokines, such as interferon-γ (IFN-γ), tumor necrosis benzoporphyrin, has been recently reported [8-10]. This
factor (TNF)-α, interleukin (IL)-12, and IL-18 [3]. This analytical tool is used here to monitor liver tissue O
2
biochemicalburstofinflammatorymediatorstargetsmul- consumptioninConA-inducedhepatitisinmice.
tiple organs, including the liver. Its outcome is a hepato-
toxicity that is characterized by mononuclear cellular Methods
infiltrationandfociofnecrosis[4]. Reagents
The potential role of NK Tcells in the pathogenesis of The Pd(II) complex of meso-tetra-(4-sulfonatophenyl)-
ConA-induced hepatitis is confirmed by the observation tetrabenzoporphyrin (Pd phosphor) was obtained from
that mice lacking NKTcells andthose treated with anti- Porphyrin Products (Logan, UT). Dactinomycin (actino-
IFN-γ antibody (to counter this essential inflammatory mycin D, MW ≈ 1255) was purchased from Merck
mediator released by NK T cells) are both resistant to (Whitehouse Station, NJ). A lyophilized powder of caspase
ConA-induced hepatitis. In addition, targeting NK T inhibitor I (zVAD-fmk, MW ≈467.5) was purchased from
cells with glucocerebroside, anaturallyoccurringglycoli- Calbiochem (La Jolla, CA). Ac-DEVD-AMC (MW ≈729.6)
pid that inhibits NK T cell proliferation, ameliorates and caspase-3 (molecular mass ≈30.5 kDa, heterodimer
ConA-induced hepatitis [5]. These findings suggest the active human recombinant) were purchased from Axxora
pathological outcome of the liver in ConA-treated ani- LLC (San Diego, CA). Glucose [anhydrous], ConA (from
mals relies on initial activation of innate immune com- Canavalia ensifomis, Jack bean, Type IV-S), bovine serum
ponents, such as NK T cells, followed by subsequent albumin (free of endotoxin and fatty acids), and remaining
activation and recruitment of effector immune cells, reagentswereboughtfromSigma-Aldrich(St.Louis,MO).
including macrophages and T-cells. We confirm here Dactinomycin solution was made fresh in dH O; its
2
that the pathology of concanavalin A-induced hepatitis concentration was determined by absorbance at 440 nm,
is dependent on activated immune cells especially NK T using an extinction coefficient of 24,450 M-1 (cid:129) cm-1 [7].
cells andthe secretion of cytokines,including IFN-γ, and The zVAD-fmk solution (2.14 mM) was made by
mice in which the cytokine is genetically deleted develop dissolving 1.0 mg of zVAD-fmk in 1.0 ml of dimethyl
amilderform ofthedisease. sulfoxide and stored at −20°C. The Ac-DEVD-AMC cas-
ConA-inducedhepatitisisassociatedwithcaspaseactiva- pase substrate was dissolved in dimethyl sulfoxide at a
tion [6]. Caspase-3, a cysteine aspartate-directed protease concentration of 6.85 mM and stored at −20°C in small
and a family member of the IL-1β-converting enzyme aliquots. Phosphate-buffered saline (PBS) with glucose
(ICE), is the key executer of apoptosis. This enzyme is (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na HPO , 1.4
2 4
involved in proteolysis of several proteins, including poly mM KH PO , and 5 mM glucose; pH 7.4) was made
2 4
(ADP ribose) polymerase; it cleaves at the C-terminal to fresh. ConA (5 mg) was suspended in 3 ml sterile PBS
Asp216 in the DEVD (asp-glu-val-asp) sequence. This (f/c = ~1.67 mg/ml); aliquots were stored at −20°C. Pd
4-amino-acidmotifhasbeenutilizedforthehighly-specific phosphor solution (2.5 mg/ml = 2 mM) was prepared in
caspase-3 substrate, Ac-DEVD-AMC (N-acetyl-DEVD-7- dH O and stored in aliquots at −20°C. NaCN solution
2
amino-4-methyl-coumarin). Caspase-3 cleaves the tetra- (1.0 M) was prepared in dH O; the pH was adjusted
2
peptide between D and AMC, releasing the fluorogenic to ~7.0 with 12 N HCl and stored at −20°C.
moiety 7-amino-4-methylcoumarin (AMC) that can be
separatedonHPLCanddetectedbyfluorescence[7]. Mice
The status of liver tissue mitochondrial O consump- Male C57Bl/6 and C57Bl/6 IFN-γ−/− mice (8-12-wk-old;
2
tion in ConA-induced hepatitis is unknown. In this weight ≈18-22 g) used in this study were maintained at
study, compound alterations in liver tissue respiration an animal facility that was in compliance with NIH
were noted early in the course of the induced hepatitis. guidelines (http://grants.nih.gov/grants/olaw/references/
The inflammatory cytokines, including IFN-γ appear to phspol.htm).ThemicewerepurchasedfromTheJackson
contribute to these deleterious changes in hepatocyte Laboratory (Bar Harbor, ME). All mice were housed in
energymetabolism. rooms maintained at 22°C with ~60% relative humidity
The effect of inflammatory bursts on hepatocyte and a 12-hr light/dark cycle. All mice had ad libitum
bioenergetics is unknown, especially in ConA-induced access to standard rodent chow and filtered water. All
hepatitis. In theory, inflammatory cytokines are expected protocols here received approval from the Animal Ethics
to enhance cellular energy metabolism; however, this Committee-United Arab Emirates University - College
phenomenon has not yet been investigated and answers ofMedicineandHealthSciences.
are needed. Thispresent study attempts to address some Forthesestudies,experimentalmiceweregivenasingle
of these queries. Measuring liver tissue respiration, using intravenous injection (12 mg/kg body weight in a total
Al-Shamsietal.BMCGastroenterology2013,13:6 Page3of13
http://www.biomedcentral.com/1471-230X/13/6
volumeof300μl)ofConAinPBSorPBSaloneinthetail by searching for 10 phosphorescence intensities >1.0 volt
vein. At 0–12 hr after the treatment, mice were eutha- (bydefault).Peakdetectionwasaccomplishedbysearching
nized by intra-peritoneal injection of 100 μl/10 g (BW) of for the highest 10 datapoints of a pulse and choosing the
urethane (using 20% solution [w/v] in 0.9% NaCl) and datapointclosesttothepulsedecaycurve[10].
blood was collected from the sino-orbital orifice; serum The phosphorescence decay rate (1/τ) was character-
was subsequently prepared by standard protocols. At ized by a single exponential; I = Ae-t/τ, where I = Pd
necropsy,liverspecimenswerecollectedforanalyses. phosphor phosphorescence intensity. The values of 1/τ
were linear withdissolvedO :1/τ=1/τo+k [O ],where
2 q 2
Livertissue 1/τ = the phosphorescence decay rate in the presence of
Liver specimens (18–30 mg/mouse) were excised using a O , 1/τo = the phosphorescence decay rate in the
2
sharp pair of scissors (MoriaVannas Wolg Spring, cat. # absence of O , and k = the second-order O quenching
2 q 2
ST15024-10, Cambridge, UK). Following protocols previ- rate constant insec-1 (cid:129)μM-1[11].
ously outlined [8,9], the specimens were immediately Liver tissue respiration was measured at 37°C in 1 ml
immersed in ice-cold oxygenated Krebs-Henseleit buffer sealed vials. Mixing was carried out with the aid of
(115mMNaCl,25mMNaHCO ,1.23mMNaH PO ,1.2 parylene-coated stirring bars. In vials sealed from air, [O ]
3 2 4 2
mM Na SO , 5.9 mM KCl, 1.25 mM CaCl , 1.18 mM decreasedlinearlywithtime,indicatingthekineticsofmito-
2 4 2
MgCl , and 6 mM glucose [pH 7.2]), weighed and then chondrial O consumption was zero-order. The rate of
2 2
placed in 1.0 ml Krebs buffer containing 0.5% fat-free respiration(k, in μM O /min) wasthus the negative ofthe
2
bovine albumin and 3 μM Pd phosphor for O measure- slope d[O ]/dt. Sodium cyanide (NaCN) inhibited respir-
2 2
ments. Unless otherwise noted, the time period between ation,confirmingthatO wasbeingconsumedinthemito-
2
specimen collection and start of O measurement was < 5 chondrialrespiratorychain(Additionalfile1:FigureS1).
2
min. Where stated, specimens were incubated in vitro at
37°C in Krebs-Henseleit solution gassed with 95% O :5% Calibrationwithβ-glucoseplusglucoseoxidase
2
CO priortoO measurements. The calibration reaction contained PBS with 3 mM Pd
2 2
Specimens were also fixed in 4% phosphate-buffered phosphor, 0.5% fat-free albumin, 50 mg/ml glucose oxi-
paraformaldehyde and embedded in paraffin wax blocks. dase and various concentrations ofβ-glucose.The values
Sections of the fixed liver pieces (of 5–7 μm thickness) of 1/t were linear with [β-glucose]; the value of k was
q
were stained with haematoxylin and examined under a the negative of the slope (k = 101.1 sec-1 (cid:129) μM-1). The
q
light microscope. Sections of the harvested liver were value of 1/τ for air-saturated solution (without glucose)
also placed in Karnovsky’s solution and processed for was 28,330 sec-1 (coefficient of variation, C = 12%) and
v
electron microscopy[8]. for O -depleted solution (with 500 mM β-glucose, 1/τ )
2 o
2,875 sec-1 (C = 1%). The high values of C for the
v v
Oxygenmeasurements air-saturated solutions were due to the lower phosphor-
A phosphorescence oxygen analyzer was used to monitor escence intensities with high [O ] (little light reaching
2
O consumptionbyliverspecimens[8,9].O detectionwas the photomultiplier tube). O concentration was calcu-
2 2 2
performed with the aid of Pd phosphor that had an lated using,1/τ=1/τo+k [O ].
q 2
absorption maximum at 625 nm and a phosphorescence Dissolved O is expressed in mm Hg, ml O /L, mg O /
2 2 2
maximum at 800 nm. Samples were exposed to light L, or mmol/L (mM). For conver-sion: A partial pressure
flashes (600 per min) from a pulsed light-emitting diode ofoxygen(PO )of1.0mmHg=0.03mlO /L;1.0mlO /
2 2 2
array with peak output at 625 nm (OTL630A-5-10-66-E, L=1.4276mgO /L;1.0mgO /L=1000/32μM.Infresh-
2 2
Opto Technology, Inc., Wheeling, IL). Emitted phosphor- waterat760mmHgand20°C,dissolved[O ]is9.1mg/L,
2
escentlightwasdetectedbyaHamamatsuphotomultiplier or 284 μM. Using a Clark electrode, the PO of our reac-
2
tube (928) after first passing it through a wide-band inter- tion mixture (PBS with 10 mMglucose, 3.0 μM Pd phos-
ferencefiltercenteredat800nm.Theamplifiedphosphor- phor and 0.5% fat-free bovine serum albumin) was 170.5
escencedecaywasdigitizedat1.0MHzbya20-MHzA/D [± 6.6] mm Hg (n = 4), or 228 [± 9] μM. The 56 mm Hg
converter(ComputerBoards,Inc.,Mansfield,MA). differencebetween[O ]infreshwaterandthePdsolution
2
A program was developed using Microsoft Visual Basic reflectstheeffectofsalinityondissolvedO .
2
6, Microsoft Access Database 2007, and Universal Library
components (Universal Library for Measurements Com- SerumIFN-γandaminotransferases
puting Devices; http://www.mccdaq.com/daq-software/ Serum IFN-γ levels were determined using a sandwich
universal-library.aspx). It allowed direct reading from the ELISA DuoSet ELISA Develop-ment (mouse IFN-γ) kit
PCI-DAS 4020/12 I/O Board (PCI-DAS 4020/12 I/O (R&D Systems, Minneapolis, MN), according to manu-
Board; http://www.mccdaq.com/ pci-data-acquisition/PCI- facturer protocols. IFN-γ concentrations (in pg/ml) were
DAS4020-12.aspx). The pulse detection was accomplished extrapolated from a standard curve generated in parallel
Al-Shamsietal.BMCGastroenterology2013,13:6 Page4of13
http://www.biomedcentral.com/1471-230X/13/6
Figure1Liverrespiration3hrafterConAinjection(immediatelypost-tissuecollectionandafterinvitroincubation).C57Bl/6wild-type
micewereinjectedeitherwithPBS(PanelA)orwith12mgConA/kg(PanelB).Twoliverspecimenswerethencollectedpermouseat3hrpost-
injection;minutezerocorrespondstothetimeofsacrificeofananimal.Foruntreatedmice,onespecimen(25mg)wasrunatt=4(immediatelypost-
collection)andone(27mg)att=83min.Forthetreatedmice,onespecimen(20mg)wasrunatt=4(immediatelypost-collection)andone(24mg)
att=49min.Thelinesshownarelinearfits.Ratesofrespiration(k,inμMO min-1(cid:129)mg-1)andR2valuesareshownatthebottomoftheruns.
c 2
Figure2Time-dependentchangesinliverrespirationinConAhepatitisinC57Bl/6mice.C57Bl/6mice(wild-type)wereinjectedwithPBS
or12mgConA/kg.LivertissuespecimenswerethencollectedattheindicatedtimepointsandprocessedforO measurements.Valuesshown
2
arethemean(±SD),withthevaluesof“n”underneatheachtimepoint.Thep-valuesat3and12hr(comparedto0hr)arealsoshown;
remainingvalueswereinsignificant.
Al-Shamsietal.BMCGastroenterology2013,13:6 Page5of13
http://www.biomedcentral.com/1471-230X/13/6
Table1Livertissuerespirationinsamplesharvested 380 nm and the emission wavelength 460 nm. Solvent A
fromConA-treatedmiceafter12hr was HPLC-grade CH CN:H O [1:3, v/v], and Solvent B
3 2
Strain Treatments k was dH O (isocratic). The Ultrasphere IP column (4.6 ×
c 2
(μMO min-1(cid:129)mg-1) 250 mm, Beckman, Brea) was operated at 25°C at 1.0 ml/
2
min(0.5ml/minofeachpump).Theruntimewas30min.
mean±SD(n) range
C57Bl\6(wild-type) PBS(300μl) 0.26±0.04(5) 0.22-0.32
Phenotypingofinfiltratingmononuclearcellsby
ConA(12mg/kg) *0.18±0.03(5) 0.13-0.20
fluorescenceactivatedcellsortinganalysis(FACS)analysis
“n”representsnumberofindependentexperiments.*Valuesignificantly
differentfromcontrolatp≤0.005. Liver fragments were placed in RPMI with 5% fetal calf
serum (FCS; Hyclone, Logan, UT) and 25 mM HEPES,
using kit-provided recombinant IFN-γ. The level of sen- and passed through 200-gauge stainless steel mesh. The
sitivity ofthekit was20pg/ml. pellets were then collected by centrifugation (~500 × g at
W
Serum alanine aminotransferase (ALT) activity was 4°Cfor10min),suspendedin40%Percol ,andlayeredon
W W
determined on a Beckman Coulter Synchron System the top of 70% Percol . The solutions were then centri-
(Brea, CA); 150 μl serum/mouse were used in a reaction fuged at ~100 × g and 25°C for 20 min. Themononuclear
that contained L-alanine, α-ketoglutarate, NADH and cells were collected from the interface, washed twice with
lactatedehydrogenase.NADHoxidation,calculatedfrom RPMI,andre-suspendedinRPMIat3×105/ml.
the change in absorbance at 340 nm, is proportional to For immunophenotyping, the fluorescent-labeled anti-
the transaminase activity and the basis for the calcula- bodies fluorescein isothiocyanate (FITC)-conjugated rat
tion of ALT activity/sample. The level of sensitivity of anti-mouse CD3, CD4, and CD45 and phycoerythrin
the assay was 5 IU/L. Serum aspartate aminotransferase (PE)-conjugated rat anti-mouse CD8, NK T, and CD11b
(AST) activity was determined as above, in a reaction were used (all were purchased from eBioscience, San
that contained L-aspartate, α-ketoglutarate, NADH, and Diego, CA). To avoidnon-specific binding, 1×106 cells/
malate dehydrogenase. The level of sensitivity of the mouse were incubated with 20% FCS for 10 min at 4°C.
assay was5IU/L. The cells were then gently washed with PBS supplemen-
ted with 5% FCS and the resultant pellet treated with
Intracellularcaspaseactivity 100 μl aliquots(from 3 μg/mlstocks) each ofa given pair
Liver specimens (40–55 mg) were incubated at 37°C in of antibodies, i.e., FITC-anti-CD4 + PE-anti-CD8, FITC-
oxygenatedKHbuffercontaining74μM Ac-DEVD-AMC anti-CD3 + PE-anti-NK/NKT, or FITC-anti-CD45 + PE-
withandwithout43μMzVAD-fmk(finalvolume=0.5ml) anti-CD11b. The cells were then incubated for 35 min at
for30min.Attheendoftheincubationperiod,thesuspen- 4°C in the dark. Thereafter, the cells were washed twice
sion was sonicated on ice for 60 sec and passed 10 times with PBS-FCS and then re-suspended at 106 cells/tube in
through a 27-gauge needle. The cleavage reaction was 0.5 ml PBS-FCS. Samples were then analyzed via flow
quenched with tissue disruption, as caspases became in- cytometry using a FACSort system (Becton Dickinson,
activeduetothedilution.Thesupernatantswerethencol- Franklin Lakes, NJ) and associated CellQuest software. A
lected by centrifugation (~12300 × g for 15 min, 25°C), minimumof10,000eventspersamplewereacquired.
separated via HPLC and analyzed for free AMC as
describedbelow[7]. Statisticalanalysis
The Student’s t-test of difference of means was used to
HPLC compare treated and untreated samples. The threshold
The analysis was performed on a Waters 1525 reversed- ofsignificance wastaken tobep<0.05.
phaseHPLCsystem(SpectraLabScientificInc,Alexandria,
VA) that consisted of a manual injector, pump, and fluor- Results
escence detector. The excitation wavelength used was HepatocyteoxygenconsumptioninC57Bl/6micetreated
withConA
Table2Livertissuerespirationintissuescollectedfrom Figure1depictsarepresentativeexperiment of hepatocyte
ConA-treatedC57Bl/6wild-typeandIFN-γ−/−mice7hr
respiration in ConA-treated and untreated C57Bl/6 mice.
aftertreatment
OnemousewasinjectedwithPBS(Panel A)andonewith
Strain Treatment kc 12 mg ConA/kg per body weight (Panel B). Two liver
(μMO2min-1(cid:129)mg-1) specimens were collected from each mouse 3 hr after
C57Bl\6 PBS(180μl) 0.27±0.13(11) injection. For theuntreated mouse, one specimen was run
C57Bl\6 ConA(12mg/kg) 0.26±0.07(3) immediately and one after ~83 min of in vitro incubation
C57Bl\6IFN-γ−/− ConA(12mg/kg) 0.25±0.10(3) at37°CinKrebs-Henseleitsolutiongassedwith95%O2:5%
“n”representsnumberofindependentexperiments.Valuesshownaremean±SD. CO2. The rate of respiration (kc, μM O2 min-1 (cid:129) mg-1) in
Al-Shamsietal.BMCGastroenterology2013,13:6 Page6of13
http://www.biomedcentral.com/1471-230X/13/6
Figure3Time-dependentchangesinliverrespirationinConAhepatitisinC57Bl/6IFN-γ−/−mice.Micewereinjectedwith20mgConA/kg
andliverspecimens(24–29mg)werecollectedforO measurementsat3,6,and12hrpost-injection.ControlmicewereinjectedwithPBS.Rates
2
ofrespiration(k,inμMO min-1(cid:129)mg-1)areshownatthebottomoftheruns.Allfourrunswereperformedononeinstrument.Oneanimalwas
c 2
sacrificedat-a-timeinthefollowingsequence:12hrpost-injection,3hrpost-injection,PBS,andthen6hrpost-injection.Minutezerocorresponds
tosacrificeofthefirstanimal.
both specimens was 0.21, confirming stability of the liver
specimens in vitro[8,9]. For the treated mouse, one speci-
men was also run immediately post collection and one
after ~49 min of in vitro incubation as above. The values
of k were 0.42 (2-fold increment) and 0.18 (similar to the
c
control) respectively. These results demonstrate acceler-
ated rate of liver tissue O consumption 3 hr after ConA
2
injection. In another experiment, the values of k for the
c
untreatedmousewere0.20att=5min(immediatelypost
specimen collection) and 0.24 at t = 83 min (after 83 min
incubation in vitro as above); the corresponding values for
thetreatedmousewere0.34att=3min(a70%increment)
and0.14att=59min.Inathirdexperiment,thevaluesof
k fortheuntreatedmousewere0.25att=5minand0.23
c
at t = 58 min; corresponding values for the treated mouse
were 0.34 at t = 5 min (36% increment) and 0.20 at t = 75
min. Thus, for untreated mice, values of k (mean ± SD)
c
were 0.22 ± 0.02 (n = 6). For treated mice, values of k for
c
specimensrunimmediatelywere0.37±0.05(n=3,a68%
increase, p = 0.015) and for specimens incubated in vitro
Figure4SerumIFN-γandALTinuntreatedandConA-treated
C57Bl/6mice.C57BL/6micewereinjectedwithPBSor12mgConA/ beforetherun,0.17±0.03(n=3,a54%less,p=0.005).
kg.Serumwascollected3,6,and12hrpost-injectionandprocessed Inatotalofeightindependentexperiments,thevaluesof
forIFN-γ.IFN-γvaluesshownaremean(±SD)ofthreedeterminations k inuntreatedmicewere0.234±0.024(n=8)andintrea-
c
foreachsample(n=3animals/sample).ALTactivitiesat0,3,6,and12
ted mice 3 hr after ConA injection (12 mg/kg) 0.300 ±
hrarefromasingleexperiment(seeResults).
0.066(n=8,p<0.05).Thevariabilityinthevaluesofk in
c
Al-Shamsietal.BMCGastroenterology2013,13:6 Page7of13
http://www.biomedcentral.com/1471-230X/13/6
Figure5(Seelegendonnextpage.)
Al-Shamsietal.BMCGastroenterology2013,13:6 Page8of13
http://www.biomedcentral.com/1471-230X/13/6
(Seefigureonpreviouspage.)
Figure5LivertissuecaspaseactivationinuntreatedandConA-treatedC57Bl/6mice.C57BL/6micewereinjectedwith(A)PBSor(B-D)12
mgConA/kg.Liverspecimenswerecollected(B)3,(C)6,or(D)12hrpost-injectionandprocessedimmediatelyforcaspaseactivity.Foreach
condition,liverspecimenswereincubatedat37°CinoxygenatedKHbufferinthepresenceof74μMAc-DEVD-AMCwith(dashedline)and
without(solidline)43μMzVAD-fmkfor30min.Thecleavagereactionwasquenchedwithtissuedisruption.Thesupernatantswereseparatedvia
HPLCandanalyzedforAc-DEVD-AMC(R,~3min)andfreeAMC(R,~18min)peaksasdescribedintheMaterialsandMethods.InPanelE,liver
t t
specimensweretreatedinvitrowith8μMdactinomycinfor60minpriortotheadditionofAc-DEVD-AMCwith(dashedline)andwithout
(solidline)zVAD-fmk.
treated mice (coefficient of variation = ~22% in ConA- 3, 6, and 12 hr post-injection. The mean k value
c
injected mice compared to ~10% in untreated mice) likely (in μM O min-1 (cid:129) mg-1) for uninjected mouse was
2
reflectedtheheterogeneityoftheliverinvolvementandthe 0.27. The corresponding k values for injected mice at
c
variationofitsprogresswithtime. 3, 6, and 12 hr post-injection were 0.19, 0.22, and
Changes in liver tissue respiration during the course of 0.22, respectively (Figure 3 and Table 2). Of note,
ConA-induced hepatitis are summarized in Figure 2. As additions of exogenous IFN-γ (10, 50 or 100 ng/mL)
discussed above, the rate of respiration (k) significantly resultedininhibitionoflivertissuerespiration(Additional
c
increased at 3 hr (p < 0.02) and decreased at 12 hr file1:FigureS3).
(p=0.005,seealsoTable1).Bycontrast,thevaluesofk at
c
1,2,and6hrafterConAinjectiondidnotsignificantlydif-
ferfromuntreatedmice(alsoseeTable2).Livertissueres- SerumIFN-γinC57Bl/6micetreatedwithConA
piration was not altered by in vitro addition of ConA Serum IFN-γ levels were determined in untreated and
(Additionalfile1:FigureS2). ConA-treated (12 mg/kg) C57Bl/6 mice at 0, 3, 6, and
12 hr post-injection. The results of three independent
HepatocyteoxygenconsumptioninC57Bl/6IFN-γ−/−mice experimentsareshowninFigure4.Relativetothe values
treatedwithConA/kg seen in untreated animals, elevated levels of IFN-γ were
C57Bl/6 IFN-γ−/− mice were injected with 20 mg noted at 3 hr after ConA injection, and these increased
ConA/kg or PBS and then liver pieces were collected with time.
A BB C
D EE F
Figure6Micrographsofhematoxylin-stainedliversectionsfromuntreatedandConA-treatedC57Bl/6andC57Bl/6IFN-γ−/−mice.
Samplesarerepresentativeofthosefrom(A,B,C)wild-typeandfrom(D,E,F)IFN-γ−/−C57Bl/6mice,eachofwhichhadbeeninjectedwith12
mgConA/kg.Liverswerethencollectedat(A,D)3,(B,E)6,and(C,F)12hrpost-injection.Notetheaccumulationofmononuclearcellsinthe
wild-typemiceat3and(moreprofusely)6hr(arrow),andthesolitarycellsintheIFN-γ−/−mice(even6hrafterinjection).Infiltrationisfound
aroundvesselsandsinusoidsinbothstrains.At12hrpost-injection,areasofnecrosis(star)filledwithmononuclearcellsbecomeapparentinthe
liverofwild-typemice,butliversofIFN-γ−/−miceremainedunaffected.Bar=50μm.
Al-Shamsietal.BMCGastroenterology2013,13:6 Page9of13
http://www.biomedcentral.com/1471-230X/13/6
A Figure7Electronmicrographsofhepatocytesfromuntreated
andConA-treatedC57Bl/6mice.Hepatocyteswereobtainedfrom
(A)normaluntreatedandConA-injectedC57Bl/6mice(B)3and(C)
6hrpost-injection.Arepresentativemicrographisshownforeach
timepoint.Notethenucleiandmitochondriaineachofthe
micrographsarenormal.Bar=0.3μm.
SerumALTandASTinC57Bl/6andC57Bl/6IFN-γ−/−mice
N
treatedwithConA
Serum ALT activity was determined in untreated and
ConA-treated (12 mg/kg) C57Bl/6 mice at 0, 3, 6, and 12
hrpost-injection.ElevatedlevelsofALTwerenotedat3hr
after ConA and these increased with time (Figure 4). In
three independent experiments, serum ALT levels 12 hr
after ConA were 440 ± 393 IU/L (mean ± SD, n = 3) and
AST768±528(n=3).
For C57Bl/6 IFN-γ−/− mice, ATL activities were 45, 96,
and32IU/L atthe0,3and6hrtimepoints,respectively.
B
ThecorrespondingvaluesforASTwere157,259,and228
IU/L. At 12 hr, the values of ALT were 172 ± 154 IU/L
(n=3)andAST394±198IU/L(n=3).
Serum ALTactivities in PBS-treated C57Bl/6 mice at 0
and 12 hr were, respectively, 30 and 30 U/L; serum IFN-
γlevelsinPBS-treatedC57Bl/6miceat0and12hrwere
non-measurable. Serum ALT activities in PBS-treated
C57Bl/6 IFN-γ−/− mice at 0 and 12 hr were, respectively,
35and46U/L.
N
CaspaseactivationinC57Bl/6micetreatedwithConA
Caspaseactivityinlivertissuewasmonitoredinuntreated
and ConA-treated (12 mg/kg) mice at 3, 6, and 12 hr
post-injection,usingthe caspase-3substrateanalogueAc-
DEVD-AMC. Liver specimens were incubated with 74
μM Ac-DEVD-AMC in the presence and absence of the
pan-caspase inhibitor zVAD-fmk. Caspases cleaved Ac-
C DEVD-AMC, releasing the fluorogenic moiety AMC;
post-sample disruption, AMC was separated on HPLC
(R ~18min)anddetectedbyfluorescence.
t
AMC peak was neither detected in the untreated
mouse (Figure 5A) nor in the ConA-treated mouse at 3
hr (Figure 5B). By contrast, an AMC peak was detected
at 6 and 12 hr after ConA injection. At 6 hr (Figure 5C),
the AMC peak area (arbitrary units) without zVAD-fmk
was 1,417,450 and with zVAD-fmk was 691,348 (~51%
inhibition). At 12 hr (Figure 5D), the AMC peak area
without zVAD-fmk was 2,028,388 and with zVAD-fmk
1,252,429 (~38% inhibition). These results demonstrate
N rapid uptake and hydrolysisofAc-DEVD-AMCby active
caspases in hepatocytes. Residual caspase activity in the
presence of zVAD-fmk was due to the simultaneous
addition of Ac-DEVD-AMC (added in excess) and
zVAD-fmk. The free AMC moieties were negligible in
experiments where Ac-DEVD-AMC was added 10 min
after addition of zVAD-fmk (data not shown). In another
Al-Shamsietal.BMCGastroenterology2013,13:6 Page10of13
http://www.biomedcentral.com/1471-230X/13/6
Figure8Electronmicrographsofsectionsofmurinelivertaken12hrafterintravenousinjectionof12mgConA/kg.Representative
samplesshownarefromC57Bl/6(B,D)andC57Bl/6IFN-γ−/−(A,C)mice.Notetheareasoffibrosis(**)andderangedmitochondria(★)inthe
liverofC57Bl/6mice.Withtheexceptionoffattyinfiltrates(▲),theliveroftheC57Bl/6IFN-γ−/−miceappearsnormalwithintactmitochondria
(m).Bars:A-B=0.5μm;C-D=0.2μm.
two independent experiments, the AMC peak was de- C57Bl/6, but not in C57Bl/6 IFN-γ−/−, mice. Overall, the
tected at 3 hr and increased linearly at 12 hr (R2 >0.999 liver architecture was relatively preserved in the C57BL/6
and>0.995). IFN-γ−/−mice.
For comparison, liver tissue specimens were also trea- Electron micrographs of the hepatocytes showed a nor-
ted in vitro with 8 μM dactinomycin, a cytotoxic agent malpresenceofnucleiandmitochondriainuntreatedand
that intercalates between DNA base pairs and activates ConA-treated(12mg/kg)C57Bl/6miceat3and6hrafter
caspases [7]. The AMC peak area in the presence of injection(Figure7).
dactinomycin alone was 6,504,578 and in the presence Thehistopathologyofliversectionsobtainedfromunin-
of dactinomycin + zVAD-fmk was 1,427,361 (78% inhib- jected or PBS-injected C57Bl/6 and from uninjected or
ition, Figure5E). PBS-injected C57Bl/6 IFN-γ−/− mice showed an absence
ofinfiltrationsinanyofthesectionsexaminedandprom-
LightandelectronmicroscopyoftissuesfromC57Bl/6 inentsinusoidsintheliversofthePBS-injectedhostsfrom
andC57Bl/6IFN-γ−/−mice bothstrains(Additionalfile1:FigureS4).
Figure 6 shows the liver histology in ConA-injected Electron micrographs of sections of murine liver taken
(12 mg/kg) C57Bl/6 and C57Bl/6 IFN-γ−/− mice at 3, 6 12 hr after intravascular injection of ConA (12 mg/kg)
and 12 hr. In C57Bl/6 mice, mononuclear cell accumula- from C57Bl/6 and C57Bl/6 IFN-γ−/− mice are shown in
tion around blood vessels was noted 3 and 6 hr after the Figure 8. Areas of fibrosis and deranged mitochondria
ConA administration. More prominent mononuclear cell were prominent in the liver of C57Bl/6 mice. With the
infiltration around blood vessels and into the tissues and exception offatty infiltrates, the liver ofC57Bl/6IFN-γ−/−
areas of necrosis were noted 12 hr after the injection in miceappearednormal-withintactmitochondria.