HAL Id: hal-02692529
https://hal.inrae.fr/hal-02692529
Submitted on 1 Jun 2020
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of
sci-entific research documents, whether they are
pub-lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Inhibitory effect of interferon-gamma activated ovine
umbilical vein endothelial cells on the intracellular
replication of Toxoplasma gondii
I.H. Dimier, D.T. Bout
To cite this version:
Original
article
Inhibitory
effect of
interferon-γ
activated ovine
umbilical vein endothelial cells
on
the intracellular
replication
of
Toxoplasma
gondii
IH Dimier*
DT Bout
CJF INSERM 93-09
Immunologie
des maladies infectieuses,Équipe
associée Inraimmunologie
parasitaire,
UFR des sciencespharmaceutiques,
31, avenueMonge,
37200 Tours, France(Received
21 December 1995; accepted 25April 1996)
Summary ― Toxoplasma gondii
is anobligate
intracellular parasite that is amajor
cause of abortionand neonatal
mortality
in sheep. Incongenital toxoplasmosis, Tgondiifirst
invades the umbilical veinendothelial cells and are then disseminated
throughout
the fetus. Treatment of ovine umbilical veinendothelial cells with bovine recombinant
y-interferon (IFN-y)
blocked thegrowth
ofT gondii.
Growth of the parasite was measuredby
3H-uracilincorporation
18 h after the onset of the infection andby
micro-scopic
enumeration ofparallel
cultures. This assay revealed that when the cells werepretreated
withIFN-y in
concentrationsranging
from 0.15-1 250 U/mL, ahigh degree
of inhibition ofT gondii
replica-tion was observed with the effect
being dose-dependent.
Maximum activation was achievedby
incu-bating
with 625 U/mLIFN-y and
noactivity
was present at 0.15 U/mL. Thistechnique
could be ofrel-evance as a first line of defense
against congenital
ovineToxoplasma
infection. Inhibition ofT gondii
replication
is due to a different mechanism from thatexisting
in mousemacrophages
and humanfibroblasts.
L
-Arginine-dependent
production of reactivenitrogen
and oxygen intermediates was notresponsible
for the inhibition ofT gondii replication. Supplements
of five amino acids were able toovercome the inhibition
partially
butsignificantly.
The mechanism of the inhibition remains to beelu-cidated.
ovine umbilical vein endothelial cell
/ Toxoplasma gondii
l IFN-y
Résumé ― Les cellules endothéliales de la veine ombilicale ovine activées par l’interféron y
inhibent la
multiplication
intracellulaire deToxoplasma gondü. Toxoplasma gondii
est un parasiteintracellulaire
obligatoire
àl’origine
de nombreux avortements et de mortalité néonatale chez lemou-ton. Dans la
toxoplasmose congénitale,
Tgondü
infecte dans unpremier
temps les cellules endothéliales*
de la veine ombilicale avant de se disséminer à l’ensemble du foetus. Les cellules endothéliales de la veine ombilicale ovine activées par l’interféron gamma
(IFN-y)
recombinant bovin inhibent lamultipli-cation de T
gondü.
Lamultiplication
duparasite
est estimée parincorporation d
3
H-uracile,
18 heuresaprès
l’infection et par numérationmicroscopique
de culturesparallèles.
Les cellulespré-incubées
avec de
l’IFN-yselon
une gamme de concentrations allant de 0.15 à 1250 Ulml montrent un fortdegré
d’inhibition deréplication
de Tgondü
avec un effetdose-dependant.
L’activation maximale est obtenueavec 625 Ulmi
d’IFN-y,
aucune activité n’est décelée à 0,15 Ulml. L’inhibition dereplication
est due àun mécanisme différent de ceux rencontrés dans les
macrophages
murins et les fibroblastes humains.La
production
de dérivés nitrés etoxygénés n’est pas responsable
de l’inhibition deréplication
de Tgondii.
L’addition de 5 acides aminés lèvepartiellement
maissignificativement
l’inhibition. Le mécan-isme d’inhibition reste inconnu.cellules endothéliales de la veine ombilicale
ovine/Toxoplasma gondii lIFN-y
INTRODUCTION
Toxoplasma
gondii
is aprotozoan
parasite
that infects all warm-bloodedanimals,
including
man(Werk, 1985).
In thefarming
industry,
ovinetoxoplasmosis
is ofgreat
economicimportance,
because it causessheep
to abortleading
to neonatal loss(Bux-ton,
1990).
WhenT gondii is
acquired by
apreviously unexposed
individualduring
preg-nancy, it can cross theplacenta
and may cause an acute infection in theimmunolog-ically-incompetent
fetus.Cell-mediated
immunity
toT gondii is
believed to be the
major
mechanism of resistance totoxoplasmosis
and is medi-atedthrough IFN-y
activation(Suzuki
andOrellana, 1988). Understanding
the mech-anisms of host resistance andidentifying
thecomponents
ofprotective immunity
is animportant prerequisite
for thedevelop-ment of novel and safe ways of
controlling
toxoplasmosis.
During congenital
toxoplasmosis,
infec-tion of the umbilical cord blood vessels is themajor
route ofpotential
transmission tothe fetus.
Experiments
with human umbilical vein endothelial cells have revealed that theanti-toxoplasma activity
of these cells can be stimulatedby
humanIFN-y (Suzuki
andRemington, 1988).
It was consideredimpor-tant to extend these results to the ovine model. In this
report
we observed thatbovine recombinant
IFN-y significantly
enhanced theability
of ovine umbilical vein endothelial cells(OUVEC)
to inhibit thegrowth
ofT gondii
and we tested the effect of oxygen intermediates andtryptophan
on this inhibition.MATERIALS AND METHODS
Cell-culture
Ovine endothelial cells were obtained from
umbil-ical cord veins
by digestion
with 0.2%collagen-ase, as was
previously
carried out in the human model(Wautier and
Wautier,1984).
OUVEC werecultured in RPMI 1640 medium
(GIBCO)
con-taining
10% fetal calf serum(Flow),
2 mML
-glu-tamin(GIBCO),
20 mMHepes (GIBCO),
100U/mL
penicillin (Sigma),
100N
g/mL streptomycin
(Sigma)
in 25-cm3flasks(Nunclon).
The cultureswere maintained at 37 °C in 5%
C0
2
at 95%humidity.
When confluence was reached, the pri-mary OUVEC cultures weretrypsinized (0.2%
trypsin
in 0.02%EDTA)
and seeded into 96-wellplates (Falcon)
at a concentration of 2.5 x 104 cells/well for the[
3
H]uracil
test or in 24-wellplates
(Falcon)
at a concentration of 5 x 104in 200pL
formicroscopic
examination.Production ofT
gondii tachyzoites
female Swiss
(OF1)
mice. For the in vitro infection of OUVEC,tachyzoites
were harvestedby
peri-toneallavage
2 or 3days
afterip
infection intomice, counted with an
haemocytometer
and added to the OUVEC culture within 3 h ofbeing
harvested at a concentration of 5 x 104protozoa
per well.
Assay ofT gondii
intracellularreplication
and effect ofIFN-y
When the OUVEC cultures were confluent, the culture medium was
replaced
with variouscon-centrations of
IFN-y ranging
from 1 250-0.15 5U/mL.
The cultures were incubated for 22 h and then washed twice.
Tgondiitachyzoites
were added to the OUVEC in 200pL
ofseeding
medium(5
x10
4
tachyzoites/well)
and theplates
were returnedto normal culture conditions. After 2 h, 2.5
N
Ci
5,6-[
3
H]uracil (Amersham, specific
activity 50 Cimmol-1)
were added to each well in 50pL
medium and the cultures were incubated for afurther 16 h. This precursor is
incorporated
by
Tgondii
but notby
the host cells becauseonly
theparasite contains
significant quantities
of therequired salvage
enzyme uracil phosphoribosyl-transferase. The incorporation of thiscompound
is correlated with intracellular
growth.
Uninfected cultures were also labeled in everyexperiment
and
consistently
incorporated <1% of theradioac-tivity
found in the control cultures that wereinfected but not treated with interferon. Prior to cell
harvesting,
the culture appearance waschecked to
verify
whetherT gondii-induced
celllysis
had started.Non-incorporated [
3
H]uracil
andtachyzoites
which could have failed to infect theOUVEC were removed by two washes with
phos-phate-buffered
saline. The OUVEC were thenlysed by
the addition of 0.1 % sodiumdodecyl
sul-phate (Serva,
100pL/well)
and, after 30 min atroom temperature, the nucleic acids were
pre-cipitated by the addition of 3 M trichloroacetic
acid
(UCB,
50pL/ well).
The contents of the wellswere
deposited
onglass
fiber filters with an auto-mated cell harvester(Skatron)
andradioactivity
was determined in an LKB liquid scintillation
counter using ACS II scintillation fluid
(Amer-sham).
The results wereexpressed
as countsper minute
(CPM).
For eachexperiment,
controls included OUVEC, eitherpre-incubated
or not withIFN-y,
cultured in the absence ofT gondii.
Theradioactivity
level for these samples was alwaysaround
background
level, thusconfirming
that[
3
H]uracil incorporation
wasspecific
to the para-site.Effect of
N
G
MMA
on theIFN-y induced
inhibition ofTgondii replication
The
protocol
for the culture, activation andinfec-tion of OUVEC was as described above except
that various concentrations of NGMMA
(p-hydroxy-azobenzene-p’-sulfonate
salt; molecularweight,
484.3;
Sigma), ranging
from 15-500pM
andfreshly
made incomplete
RPMI, were addedalong
with theIFN-y (each
in 100pL).
The finalconcentration of
IFN-y
in the cultures was100 U/mL. After incubation and
washing,
NGMMAwas
again
added to the culturesimmediately
before the addition of T
gondii tachyzoites (each
in 100pL).
Measurement of OUVEC nitrite
production
The presence of nitrite in 24-h supernatants of
IFN-y-activated
OUVEC was determinedby
theGriess reaction. Cell culture supernatant
(150 uL)
was added to 700
pL
of afreshly-prepared
mixture(50/50)
of 1°/ sulfanilamide(Sigma)
in 1.2 Nhydrochloric
acid and 0.3%N-1-naph-thylethylenediamine dihydrochloride (Sigma).
Theabsorbance at 450 nm was determined after
incu-bation for 30 min in the dark. Nitrite
concentra-tions were calculated
by
reference to a calibrationcurve
prepared by using
standard solutions ofsodium nitrite
(Prolabo)
made up in the samemedium as for the cell culture.
Effect of oxygen scavengers on the
IFN-yinduced
inhibition ofTgondii
replication
OUVEC were cultured, activated and infected as
previously described, except that
oxygen-inter-mediate scavengers were added to the cultures
together
withIFN-y (100 U/mL) during
the last 3 h of the incubationperiod
and the scavengersfinal concentrations were as follows: 2.23
mg/mL
superoxide
dismutase(3
250U/mg
0
2
-scav-enger),
1.91mg/mL
catalase(42
000U/mg,
64
mg/ml, H
2
0
2
scavenger),
50 mM D-mannitol(OH-scavenger),
1 mM DABCO(0
2
-quencher),
10 mM benzoic acid
(sodium
salt,OH-scavenger),
and 10 mM L-histidinehydrochloride
(0
2
-quencher).
All werepurchased
fromSigma
andwere made up
immediately
before use incom-plete
RPMI medium. These concentrations andconditions
successfully
blocked theoxygen-dependent
inhibition ofTgondii replication
in thestudy
performed by Murray
et al(1979).
Effect of
!-tryptophan
and other amino acids on theIFN-y-induced
inhibition ofTgondii
replication
OUVEC were cultured, activated and infected
as
previously
described, except that exogenousL
-tryptophan (Sigma)
incomplete
RPMI wasadded
along
with theIFN-y (100 U/mL)
and wasthen added
again immediately
before addition of the Tgondii tachyzoites.
The concentrationof the amino acids in the wells
ranged
from 0.5-1 000pg/mL.
Statistical
analysis
Values shown are the means of
triplicate
sam-ples
with SE bars and arerepresentative
of at least four independentexperiments.
Statistical differences between the various groups wereassessed
using
the Student’s t test. The level ofsignificance
was set at 0.05.RESULTS
Effect
of IFN-y on
Tgondii replication
in OUVECThe in vitro treatment of OUVEC with IFN-y stimulates these cells to inhibit the
growth
of the intracellularparasite
T gondii.
The treated cells become able toprevent
the otherwise unchecked intracellularreplica-tion of
T gondii.
We demonstrated thisactivity by
measuring
theincorporation
of[
3
H]uracil
by
theproliferating parasites
andthrough microscopic
observation of the infected cultures. In normal unactivatedOUVEC,
Tgondii incorporated
large
amounts of
[
3
H]uracil,
suggesting
arapid
and markedproliferation
(fig
1).
In contrast,[
3
H]uracil
incorporation
wasdramatically
reduced in OUVEC that had beenprein-cubated with
IFN-y (fig
1
). Maximal
activa-tion ofanti-Toxoplasma activity
was obtained with 625 U/mL ofIFN-y.
TheIFN-y-induced
inhibition wasdose-dependent,
with a minimum effective dose of about 0.15 U/mL(fig 1
Microscopic
observation of the infected endothelial cells confirmed the results of theisotope uptake
assay. In atypical experiment
where the cultures were treated with 150 U/mL IFN for 18h,
the number of
Toxoplasma
per endothe-lial cell was 8-16 perparasitophorous
vac-uole in the controls and 1-2 perpara-sitophorous
vacuole in the treated cultures(fig
2).
There was nosignificant
difference however between the average number of infected cells in the untreated cultures(25
±
1%)
and in theIFN-!y-treated
cultures(150 U/mL)
2 h(27
± 2%)
and 18 h(25
±3%)
after the onset ofinfection,
suggesting
that
IFN-y
had no direct effect on the Tgondii
infection. The effect ofIFN-y
treat-ment was tomarkedly
decrease the num-ber of intracellularToxoplasma
ascom-pared
to the control cultures.Effect
of N
G
MMA
on theIFN!y-induced
inhibition ofTgondii replication
and measurement of nitriteproduction
For all tested concentrations
(15-500 pM),
N
G
MMA
had no effect on theIFN-y induced
inhibition ofT gondii
replication by
OUVEC.Effect of oxygen scavengers on the
IFN-!induced
inhibition ofTgondii
i ireplication
None of the six oxygen scavengers tested
(Superoxide
dismutase, catalase,
D-manni-tol,
DABCO,
benzoic acid and L-histidinehydrochloride)
had any effect on the inhibi-tion ofTgondii replication by
IFN-y activated
cells.Effect of
L
-tryptophan
and other amino acids on theIFN-y-induced
inhibition ofT gondii
replication
Supplementation
of the culture medium withL
-tryptophan
did not overcome theIFN-y-induced inhibition of
Tgondii replication.
Inadditional
experiments,
concentrations of up to 1mg/mL
weretested,
with nolessen-ing
of the inhibition observed(fig 3).
In order to determine if the
inhibitory
mechanism ofTgondii replication
inIFN-y-treated OUVEC was
dependent
on another amino acid in the culturemedium,
theremaining
19 amino acids were testedsep-arately
as with theprotocol
used fortrypto-phan.
The results showed that five amino acids:tyrosin, cystein, glutamic
acid,
arginin
and
aspartic
acid at concentrations >100 pg/mL
partially
butsignificantly (P
<0.05)
overcame theIFN-y-induced
inhibi-tion ofT gondii replication (fig 3).
DISCUSSION
OUVEC possess a microbiostatic
capacity
and are able toprevent
the otherwise unchecked intracellularreplication
of Tgondii after IFN-yactivation. Significant
inhi-bition of
T gondii replication
was obtainedat concentrations of 0.15-1 250 U/mL. Above a concentration of 20 U/mL
replica-tion was
completely
inhibited(P< 0.05)
ascompared
with the controlcells, ie,
the par-asite infected the cell but did notmultiply.
It is
unlikely
thatIFN-y affected
theT gondii
directly
because the cells had been washed before theparasite
was added and there was nosignificant
difference between the number of infected cells in the untreated cultures and in theIFN-y-treated
cultures.IFN-y
is a multifunctionalcytokine
withspecies-specific immunomodulatory,
antivi-ral andantiprotozoal
activities,
all of whichprobably
functionby
different mechanisms.Toxoplasma
iscapable
ofinvading
andrepli-cating
innearly
all mammalian cells. Studies ofIFN-y-activated macrophages
and fibro-blasts have identified severalanti-Toxo-plasma
mechanisms. There are severalpotential
mechanismsby
which the inhibition ofreplication
ofT gondii
may have been effected. One such mechanism was found inactivated
macrophages. By
means ofblock-ing experiments
with acompetitive
inhibitor ofL
-arginine,
N
G
MMA
(Adams
etal,
1990)
it was shown to involveL
-arginine-dependent
production
of nitricoxide,
withsubsequent
conversion to nitrite and nitrateIn contrast to mouse
macrophages,
theL
-arginine-dependent production
ofnitro-gen intermediates does not seem to be a
significant
mechanism in the inhibition of Tgondii replication by
IFN-y-activated
OUVEC.The second
putative anti-Toxoplasma
mechanism
investigated
was the effect of toxic oxygen radical metabolites(Murray
etal,
1979).
Oxidative metabolites have beenimplicated
in the destruction ofT gondii on
the basis of studiesdemonstrating
that thecapacity
of activated mouseperitoneal
perox-ide is correlated with their
capacity
to kill thisparasite (Murray
etal,
1979).
Thispoten-tial antimicrobial mechanism seems
appli-cable to situations where intracellular
para-sites are killed but not to situations where the
parasites
areonly
inhibited frommulti-plying.
Reactive oxygen intermediates do not,therefore,
appear to constitute the mechanism of inhibition for Tgondii
repli-cation inIFN-y-activated
OUVEC.Finally,
we tested the mechanismby
whichIFN-y-activated
human fibroblasts inhibit Tgondii proliferation, namely
that oftryptophan
degradation.
Pfefferkorn(1984)
showed thatIFN-ystimulates
human fibrob-lasts toproduce
an enzyme,indoleamine-2,3-dioxygenase,
thatdegrades tryptophan,
thusstarving
T gondii of
this essential amino acid. We found thatsurplus
tryptophan
had no effect on theIFN-y-induced
inhibition ofT gondii by
cells,
despite testing
it up to thephysiologically-enormous
concentration of 1mg/mL. Among
the 19 amino acidstested,
five
(tyrosin, cystein, glutamic
acid,
aspartic
acid and
asparagin)
were observed topar-tially
decrease the amount of inhibition. Induction of other amino acid starvationby
synthesis
ofdegradation
enzymes washypothesized
liketryptophan degradation
mechanism.However,
this mechanism doesnot seem to be very efficient insofar as very
high
amino acid concentrations(1
mg/mL)
only partially
affect the inhibition ofparasite
replication.
Thus,
amino aciddegradation
may not feature in therepertoire
of mecha-nismsby
which activated OUVEC exert an antimicrobiostatic effect.The mechanism
by
whichIFN-y-activated
OUVEC inhibitreplication
remains to be elu-cidated. The fact that endothelial cells fromthe umbilical cord can be stimulated to
exhibit
anti-Toxoplasma activity
invites addi-tional studies to determine the in vivo rele-vance of thisphenomenon.
This,
in turn, may lead to studies aimed atblocking
the transmission ofT gondii from
mother tofetus,
thuscontributing
to theprevention
of ovinecongenital toxoplasmosis.
REFERENCES
Adams LB, Hibbs JB, Taintor RR, Krahenbuhl JL (1990)
Microbiostatic effect of murine activated
macrophages for Toxoplasma gondii. Role for syn-thesis of inorganic nitrogen oxides from L-arginine. J
Immunol 144, 2725-2730
Buxton D (1990) Ovine toxoplasmosis: a review. J R
Soc Med 83, 509-514 4
Murray HW. Juangbhanich CW, Nathan CF, Cohn ZA
(1979) Macrophage oxygen-dependent antimicro-bial activity. 11. The role of oxygen intermediates.
J Exp Med 150, 950-956
Pfefferkorn ER (1984) interferon-y blocks the growth of T gondii in human fibroblasts by inducing the host
cells to degrade tryptophan. Proc Natl Acad Sci USA 81, 908-912 2
Suzuki Y, Orellana M (1988) Interferon-y the major medi-ator of resistance against Toxoplasma gondii.
Sci-ence 240, 51-517 7
Suzuki Y, Remington J (1988) Dual regulation of
resis-tance against Toxoplasma gondii infection by Lyt-2+ and Lyt-1+, L3T4+T cells in mice. J Immunol 140,
3943-3949
Wautier MP, Wautier JL (1984) Culture de cellules endotheliales. In: Nouvelles techniques d’6tude d’hemostase et de thrombose (S Levy-Toledano,
ed), Les editions Inserm, Paris, 117-120 Werk R (1985) How does Toxoplasma gondii enter host
cells? Rev Infect Dis 7, 449-453
Woodman JP, Dimier IH, Bout DT (1991) Human
endothelial cells are activated by IFN-yto inhibit