ingekomen 1 0 apr. 2012

INGEKOMEN 1 0 APR. 2012
Intervet International bv
Wim de Körverstraat 35
P.O. Box 31
5830 AA Boxmeer
The Netherlands
mso-animal-nealth.com
tav. de heer dr H.C.M van den Akker
Postbus 1
3720 BA Bilthoven
(. MSD
Animal Health
Boxmeer, 05 april 2012
Geachte heer van den Akker,
In antwoord op uw brief van 9 maartj.l. betreffende verzoek om aanvullende informatie IIB voor
aanvraag lG12-023100 stuur ik u de volgende informatie:
1. Wat is het maximale volume van de voorkweek van het ‘seed virus’ voordat het uitverdeeld
wordt in roller bottles? Waar en op welke wijze vindt deze voorkweek plaats?
Er vindt geen voorkweek van het ‘seed virus’ plaats. Na de voorkweek van de cellen
worden, zoals beschreven in D.2, de roller bottles geleegd en wordt er eenzelfde hoeveelheid
virushoudend medium toegevoegd aan de cellen. Dit virushoudende medium wordt gemaakt
door seed virus uit de vriezer te nemen, te ontdooien en toe te voegen aan een vat met
kweekmedium. Nadat dit is gemengd, wordt uit dit vat de juiste hoeveelheid medium per roller
bottie toegevoegd. De hoeveelheid seed virus die per roller bottie wordt toegevoegd is minder
dan 1 ml van het seed virus, maar dit gebeurt verdund in het kweekmedium (met een volume
van 700 ml kweekmedium per roller bottie).
2. OnderAl.2 en D6 van uw aanvraag geeft u aan dat kleine hoeveelheden virus aan de roller
bottles worden toegevoegds en dat viruskweken zullen worden leeggegoten in het
inactiveringsvat. Betreffen beide stappen open of gesloten handelingen en wat zijn hierbij de
kritische punten? Vinden de handelingen bijvoorbeeld plaats in VK-ll? U dient deze
handelingen en eventuele andere kritische handelingen (overpompen van het virus tussen de
inactivatievaten) in de MI-lIl ruimte nader te omschrijven. Uit uw antwoord moet blijken hoe zal
worden voldaan aan voorschrift 4.2.3.2.j van de regeling GGO (vermijding vorming en
verspreiding van aerosolen en besmetting van externe oppervlakken).
Toevoegen van het virushoudend medium en het oogsten van de viruskweek door middel
van het leeggieten van de roller botties, zijn allebei aseptische, open handelingen, die
plaatsvinden onder laminaire downflow (klasse A condities) in een cleanroom met onderdruk
met HEPA-gefilterde lucht (MI-lll geclassificeerde ruimte). De ruimte is alleen te betreden door
de sluizen. De medewerkers werken in cleanroomkleding die het gehele lichaam bedekt
(inclusief hoofdkap en bril). Deze kleding blijft bij het verlaten van de ruimte in de sluis achter,
en wordt gedesinfecteerd. Materialen worden in/uit de ruimte gesluisd via een desinfectiesluis.
Het overpompen van virus tussen de inactivatievaten is een gesloten handeling Deze vaten
worden gekoppeld dmv welding. Het flowschema van het productieproces is te vinden in
,
bijlage 3 van de aanvraag.
3. U verwijst bijeen aantal van uw antwoorden naar een COGEM advies (o.a. C8.4) U wordt
verzocht dit advies mee te sturen om uw aanvraag te completeren.
U vindt het COGEM advies CGM/070724-01 in de bijlagen bij deze brief (Bijlage 1).
Trade register No. 16028015
.:. MSD
4. Onder C9.8, C9.9 en C11.3 verwijstu voor uw antwoord naar een aantal externe en interne
onderzoeken die reeds onder een ander dossier (IG nummer IG 03-089) zijn besproken. Het is
niet mogelijk voor uw antwoord te veiwijzen naar een ander IG dossier. U wordt verzocht deze
vraag alsnog volledig te beantwoorden en de relevante rapporten mee te sturen. Indien deze
rapporten vertrouwelijk zijn dient u een openbare samenvatting hiervan te leveren.
Uitgebreid onderzoek heeft aangetoond dat de replicatie kinetiek en het weefseltropisme van
het recombinant YF-WNV vergelijkbaar is met die van de YF-17D vaccinstam.
In het artikel van Johnson et al (2003, meegestuurd als bijlage 2 bij de originele aanvraag)
wordt aangetoond dat de vervanging van de prM en E genen in YF-17D niet heeft geleidt tot
een verbeterde replicatie in Aedes Aegypti (een vector van Yellow Fever). Voor het
PFU, terwijl voor YF-1 D en WNV
recombinant YF-WNV werd een titer gevonden van 1
1068
1085
PFU gevonden werd. Vergelijkend onderzoek
en
NY99 een titer van respectievelijk
tussen WNV NY99, YF-17D en het recombinant YF-WNV in verschillende muskieten species
(de natuurlijke vectoren) toonde aan dat replicatie van het recombinant YF-WNV in de
verschillende muskietenspecies zeer beperkt of zelfs afwezig was.
Onderzoek heeft aangetoond dat zelfs een hoge dosis van het recombinant YF-WNV zich
niet verspreidt in het lichaam van een paard en dat het niet uitgescheiden wordt (bijlage 2).
Hiertoe werden vijftien seronegatieve paarden intramusculair geïnoculeerd met een hoge dosis
(7.3 1og
10 PFU) YF-WNV. De viraemie en uitscheiding van het YF-WNV werd gecontroleerd
door bloedmonsters en swabs uit de neus, mond en het rectum gedurende 14 dagen volgend
op de inoculatie en op dag 14 na inoculatie zijn monsters genomen van de volgende weefsels:
pons, medulla oblongata, cervicaal en lumbaal ruggenmerg, lever, long, milt, nier,
spierweefsel, bijnier, lymfeknopen en de thymus. Het virus kon niet gereïsoleerd worden uit het
bloed, de swabs en weefselmonsters. Hieruit kan men concluderen dat het virus zich niet
verspreidt in het lichaam van een paard en dat het niet uitgescheiden wordt. Van WNV is
bekend dat het zich in paarden vooral spreidt naar neurale weefsels en (in veel mindere mate)
naar de nier en de milt (Cantile et al 2001, bijlage 4). Aangezien het recombinant YF-WNV niet
gereïsoleerd kon worden uit de verschillende weefsels bij deze vijftien paarden kan worden
geconcludeerd dat vervanging van de prM en E genen in YF-17D niet heeft geleid tot een
weefseltropisme gelijkend op dat van WNV.
Het recombinant YF-WNV is ook vergeleken met YF-1 7D in een neurovirulentie studie in
rhesus makaken, volgens de WHO voorschriften voor YF vaccins (Arroyo et al 2004, bijlage 3).
5 PFU van het
Rhesus makaken werden in deze studie intracraniaal geïnoculeerd met i0
YF-17D Het profiel
PFU
van
i0
of
)
01
ChimeriVax-WN
recombinant YF-WNV (hier genoemd
en de duur van de viraemie na vaccinatie was vergelijkbaar voor de twee virussen. De
histologische lesies in de recombinant YF-WNV groep waren mild en zelfs minder dan die in
de groep van het YF-1 7D vaccin. De distributie van de lesies was vergelijkbaar voor de twee
virussen.
De drie bovenstaande studies tonen aan dat de replicatiekinetiek en het weefseltropisme van
het recombinant YF-WNV vergelijkbaar is met de YF vaccinstam YF-17D.
5. U dient onder Cl 1.3 de stelling “de virulentie van het YF-WN chimeer virus is ook lager dan de
West Nile virus donor van prM en E” nader te onderbouwen aan de hand van relevante
literatuur.
In het eerder aangehaalde artikel van Johnson et al (2003) wordt beschreven dat het
recombinant YF-WNV duidelijk minder goed vermenigvuldigt in vector cellijnen dan de West
Nile NY99 donor en dat de virulentie in verschillende muskietenspecies sterk verminderd is
(zie hiervoor ook het antwoord op vraag 4).
Arroyo et al (2004, bijlage 3) beschrijven dat het recombinant YF-WNV significant minder
neurovirulent is dan de YF-17D vaccinstam. Een hele lage dosis van wildtype West Nile NY99
(1-4 plaque forming units, PFU) veroorzaakte al een dodelijke encephalitits bij intraperitoneale
PFU van het recombinant YF-WN (hier genoemd ChimeriVax
toediening in muizen, terwijl
) geen encephalitis veroorzaakte.
01
WN
) MSD
Deze twee onderzoeken tonen duidelijk aan dat de virulentie van het recombinant YF-WNV
lager is dan de virulentie van de West Nile virus donor.
6. U dient uw antwoord op vraag C9. 10 te onderbouwen aan de hand van relevante literatuur.
Yellow Fever virus en West Nile virus zijn beide flavivirussen met vergelijkbare fysisch
chemische eigenschappen. Beide virussen zijn gevoelig voor verhoogde temperatuur en zure
pH. Het is niet te verwachten dat het recombinant YF-WNV een verhoogde overlevingskans
heeft in het milieu ten opzichte van YF-17D. Sood et al (1993, bijlage 4) laten zien dat YF-17D
50 in 21 dagen compleet geïnactiveerd wordt bij een
met een titer van 3.7 mouseLD
temperatuur van 37°C. Onderzoek (bijlage 6) toont aan dat het recombinant YF-WNV met een
50 na 4 dagen bij 27°C niet meer aan te tonen was. Op basis hiervan
titer van 6.3 log
10 TCID
kunnen we concluderen dat het recombinant YF-WNV geen verhoogde overlevingskans heeft
in het milieu.
7. Onder 0.4 verwijst u naar validatiestudies m.b.t. de inactivatie van het virus. U wordt verzocht
een openbare versie van de validatiegegevens van deze studie mee te sturen.
De virusoogst wordt chemisch geïnactiveerd met bromo ethylamine hydrobromide (BEA).
BEA wordt omgezet tot de actieve vorm BEl door de pH boven de 7.5 te brengen. Na ± 18 24
uur incubatie bij de juiste temperatuur en pH wordt de inactivatie chemisch gestopt (met
natrium thiosulfaat).
Deze inactivatiemethode is gevalideerd voor YF-WNV in twee validatiestudies, waarvoor het
gehalte aan virus werd bepaald in twee methoden. De resultaten zijn weergegeven in
onderstaande tabel.
-
[
virus batch
normale
concentratie
artificieel
lOx verhoogde
concentratie
inactivae tijd [hj
Methode 1
0.5
1
1.5
2
pos
pos
pos
pos
pos
neg
neg
neg
neg
2.5
3
3.5
4
6
8
0
0.5
1
1.5
2
2.5
3
3.5
4
5
6
7
8
[
Methode 2
-
-
-
-
-
-
-
neg
neg
neg
-
pos
pos
pos
pos
pos
pos
pos
pos
pos
pos
-
-
-
-
-
pos
pos
pos
pos
pos
pos
pos
neg
-
-
-
Positief: virus aangetoond, negatief: geen virus aangetoond
-:
niet getest
.:. MSD
Methode 1 is kwantitatief maar relatief minder gevoelig; methode 2 is gevoeliger maar
kwalitatief. Voor aantonen van inactivatie van het virus is voor beide methoden alleen de
waarde weergegeven als positief of negatief (virus aangetoond of niet aangetoond).
Hoogachtend,
Intervet International bv
Bij lagen:
1. Vertrouwelijk COGEM advies CGM/070724-O1
2. Vertrouwelijk rapport: Disseminatie studie YF-WNV in paarden
3. Arroyo etal (2004) J. Virol. 12497-12507
4. Cantile et al (2001)
5. Sood etal (1993)
6. Vertrouwelijk rapport: The effect of several desinfectants on the survivability of YF-WN
chimera
Cc:
5
L
4j1
çoi’N
• coG
POSTADRES:
POSTBUS 578
3720 AN 8ILTHOVEN
Aan de minister van
Volkshuisvesting, Ruimtelijke
Ordening en Milieubeheer
Mevrouw dr. J.M. Cramer
Postbus 30945
2500 GX Den Haag
DATUM
KENMERK
ONDERWERP
TEL.: 030 274 2777
FAX: 030 274 4476
[email protected]
WWW.COGEM.NET
24juli 2007
CGM/07072401
Advies recombinant YellowFever virus vaccin met West Nile virus insert
Geachte mevrouw Cramer,
Naar aanleiding van een adviesvraag betreffende een verzoek tot wijziging in de
vergunning IG-03-089103 met de titel ‘Evahiatie van een Yellow Fever virus (YFV)
recombinant met MJE-geninsertie van het West Nile virus (WNV) in proefdieren” van
Intervet International B.V., deelt de COGEM u het volgende mee.
Samenvatting:
De COGEM is gevraagd te adviseren over een verzoek tot omlaagschaling van
experimenten met een recombinant Yellow Fever virus (YFV) vaccin. Het doel van de
studie is de ontwikkeling van een vaccin voor paarden tegen het West Nile virus
(WNV). In dit onderzoek wordt een recombinant Yellow Fever virus vaccin (YFV)
met oppervlakte eiwitten van het WNV getest in proefdieren.
In 2003 adviseerde de COGEM het chimere virus YF-WNV te beschouwen als een
volvirulent virus van pathogeniteitsklasse 3 omdat er destijds onvoldoende gegevens
waren over mogelijke veranderingen in tropisme, replicatie en de genetische stabiliteit
van het YF-WNV. Op basis van nieuwe gegevens dient de aanvrager een verzoek in
tot verlaging van het inperkingsniveau van de experimenten van ML-III naar ML-JJ
niveau.
De COGEM is van mening dat de aanvullende gegevens voldoende aantonen dat het
recombinante YF-WNV virus tenminste net zo geattenueerd is als de YF-17D
ouderstam die al geruime tijd als vaccin wordt toegepast en is ingedeeld in
pathogeniteitsklasse 2. De aanvullende gegevens hebben ook aangetoond dat er geen
veranderingen optreden in weefseltropisme of replicatiekinetiek van het YF-WNV.
Bovendien acht de COGEM de kans verwaarloosbaar klein dat hët virus zich kan
verspreiden via muggen.
De laboratoriumwerkzaamheden kunnen naar het inzicht van de COGEM veilig
worden uitgevoerd op ML-II niveau met inachtneming van de voorgestelde
aanvullende voorschriften. De COGEM is van mening dat de veiligheid voor mens en
milieu middels deze indeling voldoende gewaarborgd blijft.
Dit advies beyat vertrouwelijke informatie en zal derhalve niet gepubliceerd of op een
andere wijze openbaar gemaakt worden. De door de COGEM gehanteerde overwegingen
en het hieruit voortvloeiende advies treft u hierbij aa als bijlage.
Hoogachtend,
Prof. dr. ir. Bastiaan C.J. Zoeteman
Voorzitter COGEM
c.c.
Dr. ir. B.P. Loos
Dr. T. van der Leij
Evaluatie van een recombinant Yeliow Fever virus vaccin
COGE1VI advies CGM/070724-Ol
Inleiding
De COGEM is gevraagd te adviseren over een wijziging in de vergunning voor
werkzaamheden met het Yellow Fever virus (YFV) en het West Nile virus (WNV) van
Intervet International B.V. (IG-03-089/03). Het doel van deze studie is de ontwikkeling
van een vaccin voor paarden tegen het WNV. Hiertoe wordt een recombinant YFV met
oppervlakte eiwitten M/E-geninsertie) van het WNV getest in proefdieren.
In 2003 adviseerde de COGEM het recombinante YF-WNV vaccin te beschouwen als
een volvirulent virus van pathogeniteitsklasse 3 omdat er destijds onvoldoende gegevens
waren over mogelijke veranderingen in tropisme en rcplicatiekinetiek. Daarnaast
ontbraken gegevens over de genetische stabiliteit van het recombinante YF-WNV. Op
basis van nieuwe gegevens dient de aanvrager een verzoek in tot omlaagschaling van het
inperkingsniveau van de laboratoriumhandelingen met een recombinant YF-WNV vaccin
van ML-III naar ML-II niveau.
Vertrouweljkheid
De aanvullende gegevens betreffen onder andere een aantal studies van Intervet B.V. die
vertrouwelijke gegevens bevatten. Het YF-WNV vaccin is op basis van deze
vertrouwelijke data geregistreerd bij de APHIS/USDA en toegelaten als levend vaccin in
de Verenigde Staten. De COGEM acht deze vertrouwelijke gegevens van belang om haar
advies te onderbouwen. Daarom zal dit advies niet gepubliceerd of op andere wijze
openbaar gemaakt worden.
Yellow Fever virus (YFV,)
Het Yellow Fever virus (YFV) is een positief strengs RNA virus dat behoort tot de familie
van Flaviridae, genus Flavivirus (1). Het virus wordt verspreid via muggen en
veroorzaakt gele koorts, een tropische infectieziekte die voorkomt onder apen in Afrika
ten zuiden van de Sahara en in Zuid-Amerika. YFV kan ook worden overgebracht op
mensen door de mug Aedes aegyptii. Symptomen van gele koorts bij de mens zijn koorts,
hoofdpijn, rugpijn, misselijkheid, geelzucht en stoornissen in de nierfunctie. Daarbij
kunnen ook bloedingen in de mond en in de darmen optreden, Gele koorts is een emstige
ziekte die in ongeveer 60% van de gevallen een dodelijke afloop heeft (3). Een antiviraal
middel tegen de ziekte is niet beschikbaar. Een effectieve langdurige bescherming is
echter wel mogelijk door middel van een vaccinatie met de geattenueerde YF-17D stam.
Deze YFV stam is verzwakt door een serie mutaties die verspreid liggen over het gehele
COGEM advies CGM/070724-O1
1
genoom van het virus (2, 3, 4). De YF-17D stam is 65 jaar geleden ontwikkeld en meer
dan 350 miljoen mensen zijn er inmiddels mee gevaccineerd (5).
West Nile virus (WNV)
Ook het West Nile virus behoort tot de familie van de Flaviridae. Dit positief strengs
RNA virus werd in 1937 voor het eerst ontdekt in Oeganda en daarna werd het ook
gesignaleerd in andere delen van Afrika, West-Azië, Oost-Europa en het Midden Oosten.
In 2002 is er een uitbraak van het virus geweest in New York en sindsdien verspreidt cle
ziekte zich verder in de Verenigde Staten. Het virus komt voor bij mensen, vogels en
andere gewervelde dieren zoals katten, paarden, vleennuizen, wangzakeekhoorns en
konijnen. Verspreiding van het virus vindt voornamelijk plaats via muggen (c’ulex spp.)
terwijl trekvogels een van de belangrijkste reservoirs vormen (6).
Een infectie met WNV veroorzaakt symptomen als lichte koorts, hoofdpijn, pijn,
huiduitslag en gezwollen lymfeklieren. In enkele gevallen komt het virus in de hersenen
terecht en kan het lethale encefalitis veroorzaken. Voornamelijk mensen boven de vijftig
lopen hierop een verhoogd risico. Een vaccin of effectieve profylaxe tegen het WNV is
niet beschikbaar.
Voorgenomen werkzaamheden
De aanvrager voert experimenten uit waarbij gebruik wordt gemaakt van de humane
geattenueerde vaccinstam YF- 1 7D. In deze stam zijn vervolgens de structurele genen
prM en E uitgewisseld met de overeenkomstige genen van het WNV. De prM en E genen
codereri voor oppervlakte eiwitten die een essentiële rol spelen bij de immuniteit tegen
het WNV (7,8,9). Het recombinante vaccin YF-WNV is in de Verenigde Staten
ontwikkeld, getest en vervaardigd. In Nederland zal het vaccin volgens Europese
richtlijnen getest worden op veiligheid en werkzaamheid in muizen en paarden. De dieren
worden ingeënt met het recombinante virus YF-WNV en vervolgens klinisch
geobserveerd waarbij gekeken wordt naar ziekteverschijnselen en voornamelijk
encefalitis. Na vier tot zes weken zullen de gevaccincerdc dieren geïnfecteerd worden
met een virus isolaat van het WNV om de beschermende werking van het vaccin aan te
tonen.
Eerdere COGEM adviezen
In 2003 heeft de COGEM een advies uitgebracht over bovengenoemde werkzaamheden.
YFV en WNV zijn beide virussen van pathogeniteitsklasse 3, terwijl de geattenueerde
vaccinstam YF-17D een klasse lager wordt ingeschaald. De COGEM was van mening dat
er onvoldoende gegevens waren aangeleverd over de mogelijke effecten van de insertie
van cle oppervlakte eiwitten van het WNV in de YF-17D stam. Gegevens over cle
COGEM advies CGi’vI/070724-O1
2
genetische stabiliteit, weefseltropisme en replicatie ontbraken waardoor geen uitsluitsel
gegeven kon worden dat de insertie mogelijk een verandering van tropisme en replicatie
tot gevolg zou kunnen hebben. De COGEM adviseerde daarom destijds de voorgenomen
werkzaamheden uit te voeren op ML-III niveau.
Aanvullende aangeleverde gegevens
Naar aanleiding van nieuwe gegevens die betrekking hebben op genetische stabiliteit,
weefseltropisme en replicatiekinetiek verzoekt de aanvrager tot omlaagschaling van de
laboratoriumwerkzaamheden naar ML-II niveau.
De aanvrager verwijst naar een aantal studies die betrekking hebben op het
replicatiepatroon en weefseltropisme van de recombinante YF-WNV vaccinstam. Het
gaat om twee studies van Intervet International B.V. De eerste studie is uitgevoerd in
Florida (VS) waarbij paarden ingespoten zijn met een hoge dosis YF-WNV waarna de
viraernie en uitscheiding van het virus werd gecontroleerd. Histologisch onderzoek
toonde aan dat het virus zich niet verspreid had naar andere weefsels. Bovendien kon het
virus niet gereïsoleerd worden uit bloed, secreta monsters of weefselmonsters van de
paarden. Hieruit wordt geconcludeerd dat de vervanging van de prM en E genen in YF
17D niet heeft geleidt tot een veranderd weefseltropisme. Een tweede studie van Intervet
International B.V. in muizen en de eerder aangeleverde studies in apen en vogels bij de
aanvraag in 2003 leidt tot dezelfde conclusie.
Daarnaast presenteert de aanvrager de resultaten van een studie naar de genetische
stabiliteit van het YF-WNV virus gedurende vijf passages in celcultuur. Met latere
passages dan 5 zal in dit onderzoek niet gewerkt worden. De originele studies zijn niet
toegevoegd aan de aanvullende infonnatie. De aanvrager heeft dc genotypische stabiliteit
van het YF-WNV getest door middel van een sequentie analyse op passage 0 en 5. De
sequentie van het WNV-insert alsmede de flankerende regio’s zijn vergeleken op passage
0 en 5 en leverden een homologie van 99,7% op. Van vijf van de 2128 nucleotiden in het
genoom van passage 0 kon de identiteit niet vastgesteld worden, waardoor het verschil
van 0,3% ontstond. De sequentie van het virus vaccin na passage vijf is daarnaast
vergeleken met de sequentie van het originele construct zoals dat door Acambis mc.
ontwikkeld is. Deze vergelijking leverde een homologie van 100% op, waaruit
geconcludeerd wordt door de aanvrager dat de sequentie stabiel is gebleven gedurende de
5 passages. De genetische stabiliteit bij in vivo scriölc passage kon niet worden getest
omdat het niet mogelijk bleek het virus te reïsoleren uit monsters van paarden.
Naast de genotypische stabiliteit heeft de aanvrager een studie uitgevoerd naar de
fcnotypische stabiliteit van het recombinante YF-WNV. Om de fenotypische stabiliteit te
COGEM advies CGM/0 70 724-01
3
testen is een aantal virus plaques van het envclop eiwit (E) van WNV vergele
ken in
passage 0 en 5 met behulp van een monoclonaal antilichaarn specifiek voor E van
WNV.
Ter controle werd de vaccinstam YF-17D meegenomen. Uit deze studie blijkt dat
de mate
van expressie van E nagenoeg hetzelfde is voor passage niveaus 0 en +5 en
dat het
recombinant YF-WNV fenotypisch stabiel is gedurende deze passages.
De COGEM merkt bij de fenotypische analyse overigens op dat dit alleen aantoo
nt dat in
de onderzochte passages het E-eiwit van WNV tot expressie komt, en meer specifi
ek de
epitoop die door het gebruikte antilichaam wordt herkend. Fenotypische verand
eringen in
het E-eiwit die anders zijn dan in de betreffende epitoop kunnen door middel
van deze
test niet gedetecteerd worden.
Overweging en advies
De COGEM is van mening dat de aangeleverde gegevens voldoende aantonen dat
YF
WNV in apen, muizen en paarden net zo geattenueerd is als dc vaccinstarn YF-17
D. Uit
een artikel van Johnson (8) blijkt bovendien dat de recombinante vaccinstam YF-W
NV
slechts zeer beperkt in staat is tot replicatie (na intrathoracale inoculatie) in
bepaalde
soorten muggen, die als mogelijke vector voor het virus zouden kunnen dienen.
Orale
infectie met het recombinant YF-WNV was niet succesvol. Hieruit conclu
deert de
COGEM dat de kans op verspreiding van YF-WNV via muggen verwaarloosbaar
klein is.
De originele studies naar de genetische stabiliteit van het YF-WNV zijn niet bijgele
verd
in de aanvraag. De aanvrager presenteert uitsluitend de resultaten van deze studie.
Hoewel de COGEM van mening is dat de presentatie van de gegevens ten aanzien van
de
genetische stabiliteit enkele onduidelijkheden bevat, concludeert zij op basis
van deze
gegevens dat het recoinbinante YF-WNV genetisch stabiel is gedurende vijf passag
es in
celcultuur. De aanvrager geeft aan dat een vijftal nucleotiden in passage
0 niet
identificeerbaar bleek. Deze mogelijke puntmutaties bleken echter in de YF-17
D insert te
zitten en niet in de backbone, die de pathogeniteit en verspreidingskarakteristieke
n
bepaald. Daarnaast leverde een vergelijking van het vinisconstruct na passage
5 met het
originele construct dat door Acainbis mc. ontwikkeld is 100% homologie op. Derhal
ve
acht de COGEM het aannemelijk dat het recombinante YF-WNV genotypisch stabiel
is
gedurende de vijf passages.
De COGEM wijst er bovendien op dat het YF-WNV vaccin sinds 2006 geregistreerd
is
bij de APHIS/USDA en toegelaten is als levend vaccin in dc Verenigde Staten.
Conclusie
De oudervirussen YF-17D en WNV kunnen zich niet via aërosolen verspreiden
en de
COGEM acht het niet aannemelijk dat het recombinante virus dit wel zou kunnen
. De
COGEM advies CGM/O7O724O1
4
aangeleverde studies tonen bovendien aan dat het recombinante virus tenminste net zo
geattenueerd is als de YF-17D ouderstain die al genume tijd als vaccin wordt toegepast
en dat het YF-WNV slecht repliceert in zowel zoogdieren als vogels. Indien een
medewerker ten gevolge van een prik accident zichzelf zou infecteren met de vaccinstam
YF-WNV,
acht de
COGEM
de
kans op het ontstaan van een ziektebeeld
verwaarloosbaar idem. Dierexperirnenten in apen tonen aan dat er geen ziektebeeld
ontstaat ten gevolge van injectie met de YF-17D of de YF-WNV vaccinstam. Uit het
artikel van Johnson (8) blijkt bovendien dat de kans dat het virus zich vervolgens verder
zou kunnen verspreiden via muggen verwaarloosbaar klein is.
De laboratoriuinwerkzaaniheden kumen naar het inzicht van de COGEM veilig worden
uitgevoerd op ML-II niveau met inachtneming van de voorgestelde aanvullende
voorschriften, te weten het dragen van handschoenen en het uitvoeren van open
handelingen in een veiligheidskabinet klasse II. De COGEM is van mening dat de
veiligheid voor mens en milieu middels deze indeling voldoende gewaarborgd blijft.
Referenties
1. Van Regenmortel M.HV. (2000). Seventh report of the international committee on taxonomy
of viruses. Academic Press, San Diego
2. Arya S.C. (2002). Yellow fever vaccine safety: a reality or a rnyth? Vaccine 20, 3627-8.
3. Monath T.P. (2001). YeIlow fever: an update. Lancet Jnfectious Disease 1, 11-20
4. dos Santos C.N, Post P.R, Carvalho R, Ferreira 1.1, Rice C.M. and Galler R. (1995). Complete
nucleotide sequence of yellow fever virus vaccine strains 17DD and 17D-213. Virus
Research 35, 35-41
5.
6.
7.
8.
9.
CDC. Internet: www.cdc.gov/ncidod/dvbid/yellowfever/iridex.htrn (luli 2007)
Grant L. et al (2002). West Nile virus. Lancet Infectious Disease 2, 5 19-529
Arroyo J, Miller CA, Catalan J, and Monath T.P. (2001). YelIow fever vector live-virus
vaccines: West Nile virus vaccine development. Trends in Molecular Medicine 7, 350-4
Johnson B.W, Chambers T.V, Crabtree M.B, Arroyo 3, Monath T.P, and Miller B.R. (2003).
Growth characteristics of the veterinary vaccine candidate ChirneriVax trade mark -West Nile
(WN) virus in Aedes and Culex mosquitoes, Medical and Veterinary Entornology 17, 235-43
Langevin S.A, dyo J, Monath T.P, and Kornar N. (2003). 1-lost-range restriction of
chimeric yellow fever-West Nile vaccine in fïsh crows (Corvus ossifragus). American Journal
of Tropical Medicine and Hygiene 69, 78-80
COGEM cjdi’ies CGivf/070724-O1
5
b
J ournal Qf
Virology
/y
ChimeriVax-West Nile Virus
Live-Attenuated Vaccine:
Evaluation of Safety, Immunogenicity, and
Efficacy
Juan Arroyo, Chuck Miller, John Catalan, Gwendolyn A.
Myers, Marion S. Ratterree, Dennis W. Trent and Thomas P.
Monath
J. Virol. 2004, 78(22):12497. DCI:
10.1 128/JVI.78.22.12497-12507.2004.
0
0
CD
0
-t,
-I
0
3
Updated information and services can be found at:
http://jvi.asm.org/content/78/22/12497
These inciude:
REFERENCES
CONTENT ALERTS
.
1:
.
This article cites 35 articles, 16 of which can be accessed free
at: http://jvi.asm.org/contentl78/22/1 2497#ref-list-1
Receive: RSS Feeds, eTOCs, free email alerts (when new
articles cite this article), more»
r\)
Q
0)
-1
0
1
Information about commercial reprint orders: http://jvi.asm.org/sitelmisdreprints.xhtml
To subscribe to to another ASM Journal go to: http:Iljournals.asm.orglsitelsubscriptionsl
Journals.ASM.org
Vol. 78, No. 22
JOURNAL OF VIROL0Gy, Nov. 2004, p. 12497—12507
0022-538X/04/$08.00+0 DOl: 10.1 128/JVI.78.22.12497—12507.2004
2004, American Society for Microbiology. AJI Rights Reserved.
Copyright
ChimeriVax-West Nile Virus Live-Attenuated Vaccine: Preclinical
Evaluation of Safety, Immunogenicity, and Efficacy
Juan Arroyo,
—
Chuck Miller, John Catalan, Gwendolyn A. Myers, Marion S. Ratterree,
4 and Thomas P. Monathl*
Dennis W. Trent,”
Aca,nbis, mc., Czmbiidge, Massachusetts’; DynPort Vaccine Co, LLC, Frederick, Maryland
; Tulane National
2
4
; and Vaxin, mc., Birmingham, Alaba,na
3
Prirnate Research Center, Covington, Louisiana
0
0
Received 31 March 2004/Accepted 9 July 2004
The availability of ChimeriVax vaccine technology for delivery of flavivirus protective antigens at the time
West Nile (WN) virus was first detected in North America in 1999 contributed to the rapid development of the
vaccine candidate against WN virus described here. ChimeriVax-Japanese encephalitis (JE), the first live-at
tenuated vaccine developed with this technology has successfully undergone phase 1 and II clinical trials. The
ChimeriVax technology utilizes yellow fever virus (YF) 17D vaccine strain capsid and nonstructural genes to
deliver the envelope gene of other ilaviviruses as live-attenuated chimeric viruses. Animo acid sequence homology
between the envelope protein (E) of JE and WN viruses facilitated targeting attenuating mutation sites to develop
the WN vaccine. Here we discuss preclinical studies with the ChiineriVax-WN virus in mice and macaques.
ChimeriVax-WN virus vaccine is less neurovirulent than the commercial YF 17D vaccine in mice and nonhu
man primates. Attenuation of the virus is determined by the chimeric nature of the construct containing attenu
ating mutations in the YF liD virus backbone and three point mutations introduced to alter residues 107, 316,
and 440 in the WN virus E protein gene. The safety, immunogenicity, and efficacy of the ChimeriVax-WN
02 vaccine
in the macaque model indicate the vaccine candidate is expected to be safe and immunogenic for humans.
0
til
0
0
0
-
1
0
3
(1)
3
0
1
CD
0
Following isolation of West Nile (WN) virus in New York in
1999, the virus rapidly spread across North America, causing
disease in wild birds, horses, and humans. The number of
human cases increased dramatically in 2002 and 2003, when
4,145 and 8,977 cases were reported, respectively (7, 8). WN
virus is transmitted principally between wild birds and Culcr
mosquitoes (7). Recently, WN virus has been isolated in the
West Indies and serosurveys have identified neutralizing anti
body-positive avian species in Mexico (14), Jamaica, and the
Dominican Republic (13, 17). The rapid geographic expansion
of the virus is attributed to movement by viremic birds dunng
local and migratory flight behavior. To date, there is no effec
tive drug treatment against WN virus infection and surveil
lance and mosquito control measures have not significantly
influenced the number of human infections (27). A vaccine
against WN virus represents an important approach to the
prevention and control of this emerging disease.
The ChimeriVax technology has been successfully used to
develop a live vaccine against Japanese encephalitis (JE) virus
that is now in phase II trials (23). JE virus is a close genetic
relative of WN virus (31), a fact that expedited use of this
technology to develop multiple WN virus vaccine candidates.
The ChimeriVax technology employs the yellow fever (YF)
17D vaccine capsid and nonstructural genes to deliver the
envelope genes (prM and E) of other flaviviruses. In the work
presented here, the envelope genes of YF 17D were replaced
with the correspondirig genes of the wild-type WN virus NY99
strain previously described by Lanciotti et al. (19). The result
ing YF/WN chimera lacked the mouse neuroinvasive property
of WN virus and is less neurovirulent than YF 17D vaccine in
*
Corresponding author. Mailing address: Acambis, mc., 38 Sidney
St., Camhridge, MA 02139. Phone: (617) 761-4200. Fax: (617) 4941741. E-mail: [email protected].
both mouse and monkey models. Because WN virus, like other
flaviviruses in the genus, is neurotropic for mammals (21, 29),
attenuating point mutations were later introduced in the en
velope of the YF/WN chimera to further reduce its virulence.
Mutation sites were targeted only to regions of the envelope
(E) protein gene and were based on previous observations by
others (1, 3, 28, 32) pertaining to attenuation phenotypes in
related flaviviruses: specifically JE and tick-horne encephali
tis viruses. Site-directed mutations in the WN virus E gene
,
01
of the chimeric prototype vaccine, ChimeriVax-West Ni1e
) resulted in a significant reduction in virus
01
(ChimeriVax-WN
neurovirulence. Here we discuss a vaccine in a YF vaccine back
bone; the WN virus envelope (E) protein mutagenesis ratio
nale; and the assessment of the safety, immunogenicity, effi
cacy, and genetic stability of these ChimeriVax-WN vaccine
candidates in the mouse and macaque models.
-D
1
(ii
t’)
Q
t’)
0
CI)
-1
0
0
0
MATERIALS AND METHODS
YF/WN chimeric clones and molecular procedures For virus assembly. Chi
mciie flaviviruses were constructed with the ChimeriVax two-plaamid technoloey
previously deseribed (9). Briefly. the two-plasmid system provides plasmid sta
bility in E3chcrichia coli by dividing the cloned YF hackbone into two plasmids.
This provides smaller plasmids that are more stable to manipulate the YF
sequences faciitating replacement of the prM and E genes of the flavivirus target
vaccine. The VN virus prM and E genes used were cloned from the WN da
mingo isolate 383-99 sequence (GenBank accession no. AF196835: kindly provid
cd by John Roehrig, Centers for Discase Control and Prevention, Fort Collins,
Colo.). Virus prME sequence cDNA was ohtained by reverse transcription-PCR
(RT-PCR) (XL-PCR kit; Applied Biosystems, Foster City, Calif.). The 5’ end of
the WN virus prM gene was cloned precisely St the 3’ end of the YF 17D capsid
gene hy ovcrlap-extension PCR using Pwo polymerase (Roche Applied Science,
Indianapolis, md.). This cloning step maintaincd intcgrity of the cleavage/pro
cessing signal encoded at the 3’ end of the YF capaid gene. The 3’ end of the E
gcne was cloned at the 5’ end of the YF NSI protein coding sequence by overlap
extension PCR. The two-plasmid system used to done the prME rcgion of WN
virus into the YF 17D hackboue was deseribed previously (4). Silent mutations
were introduced in the sequences of prM and E to create unique BspEi and Eagl
restriction sites. Digestion of the two plasmids with these restriction nucleases
12497
0
1
m
0
0
1
>
m
12498
ARROYO ET AL.
J. Vttot.
TABLE 1. Switch oligonucleotides used for site mutagenesis
E protein position and residue’
Primer”
107L-sF
138E-sK
176V
176Y
280K-sM
316A-W
440K-R
5’-CAACGGCrGCGGArlrlll GGCAAAGGATCCATFGACACATGCGCC-3’
5’-GAAAGAGAATATFAAGTACAAAGTGGCCAiilu 1GTCC-3’
5’-GCCCFCGAGCGGCCGATrCAGCATCACTCCrGcrGCGCCrrCAGTCACAC-3’
5’-GCCCTCGAGCGGCCGATFCAGCATCAC-3’
5’-GCAACACTGTCATGTUAACGTCGGGTCA1TFG-3’
5’-CITGGGACUCCCGTGGACACCGGTCACGGCAC-3’
5’-GGGGTGTlCACTAGTGGTfGGGCGGGCTGTCCATCAAGTG-3’
c
Marker
site
BamHi
Sspl
Hpal
Agel
Spel
Primers indicated with an asterisk are clonirtg primers used in fragment subcloning. One incorporates a change to valine as indicated.
b
Primers for aitc-directcd mutagenesis to create YF/WN chimeric viruscs. Nuclcotidc changes that switch to a new amino acid are indicatcd in bold. Silent restriction
(marker) aites introduced are underlined.
C
C
generated DNA fragments that were gel purifled and ligated in vitro to produce
a fuil-length chimeric cDNA. The cDNA was iincarizcd with XhoI to facilitate in
vitro transcription by SP6 polymerasc (Epicentre, Madison, Wis.).
Point mutations werc introduced into various E gene codons to produce
variants of the original chimera coding for wild-type WN virus prME genes
(Transformcr site-dircctcd mutagenesis kit; Ciontech, Psio Alto, Calif.). Table 1
showa the mutation target sitcs and the oligonucleotide sequences used to create
all of the YF/WN chimeras. Site mutations were confirmed by sequencing of the
envelope proteins (prME region) of the resulting viruses. Virus cDNA tempiates
for sequencing originated from RNA cxtraction of virus containing infected Vero
ccli supernatants (Trizol IS; Invitrogen, Carisbad, Calif.) foliowed by RT-PCR
(XL-PCR kit; Appiicd Biosystems) and sequencing with a CEQ 2000XL nucleic
acid sequencer (Beckman-Coulter, Fuilerton, Calif.).
Viruses and ccli lines. The wild-type WN virus used in animai chalienge
atudies is the NY99 strain (NY99-35262-11 flamingo isolate, a homolog of the
virus uaed to build chimeras) ohtained from the Centers for Disease Control and
Prevention, Fort Coilins, Colo. (CDC stock dcsignatiors BS2332W) with two
additional passages in Vero E6 cella to produce a master virus bank. YF 17D is
a commercial vaccine (YF-VAX; Aventis Pasteur, Swiftwatcr, Pa.) used after
rcconstitution of the lyopliiiized product or after one passage (Pl) in Vero E6
(Amcrican Type Cuiture Coliection [ATCC] origin; Acambis, me., ccli bank,
Cambridge, Masa.). Chimerie YF/WN (ie., ChimcriVax-WN) viruses wcrc prc
pared by RNA transfection (Pl virus) of Vcro E6 ceils (ATCC origin, CtDVR
University of Massachusetts Medical Conter ccii bank, Worccstcr, Mass.). Re
search master seeds (RMS) werc prcparcd by additional amplifications (either
passage 2 or 3 St a 0.001 multiplicity of infection IMOI]) in Vero E6 celis. Vero
E6 celia were maintained in minimal esscntial medium (Invitrogen) containing
10% heat-inactivated fetal hovine serum (Hl-FBS) (HyClone, Logan, Utah).
Preparation of pre-master seeds (PMS) for manufacture of the vaccine was
initiated hy RNA transfcction of semm-free Vero (SF-Vero) celis obtained from
a ccli bank that had heen manufactured and controiled to meet current Food and
Drug Adminiatration guidelines for ccli cuiture vaccines. (The celis were oh
tained from an ATCC strain predating 1980, and the ccli bank was made by
Baster/Immuno, Orth, Austria.) Progeny virus from the transfection step was
amplifled by a single passage in the same SF-Vero cell line to produce P2, which
was designated the PMS for subsequent manufacture of clirtical-grade vsccine.
The SF-Vcro ccli line is propagated and maintsined in a serum-free, animal
protein-free medium formulation, VT-Media (Baxter/Immuno). Viruses for an
imal cxperiments were diluted in Mt99 with HEPES buffer (Invitrogen) and 20%
Hl-FBS (HyClone) uniess otherwise indicated. Piaquc sssays to vcrify the titer of
virus inoculi were performed in a Vero ccli substrate as previousiy described (24).
Mouse studies. Protocois for mouse experiments were approved by the Insti
tutional Animal Care and Use Committees at both University of Massachusetts
Medical Center (Worcester, Mass.) and Acambis, mc. (Cambridge, Mass.). Re
search was conducted in compliance with the Animal Welfare Act and other
federal statutes and regulations relating to animais and experiments invoiving
animals and adhered to principlea set forth in the Guide for ihe Care and Use of
Lahoralo,y AnirnaLs (27a). Female 1CR mice (Taconic, Germantown, N.Y., or
Harlan Sprague-Dawiey, Indianapolis, md.) wcrc inoculated intraperitoncally (i.p.)
with 100 to 200 pi of wild-type WN virus NY99 for neuroinvasion tests or post
vaccination chailenge experimenta (titers of inoculated viruscs are indicatcd in the
Results section and in the tabies presented). 1CR strain aduit (3 to 4 weeks of age)
and suckling (2 and 8 days of age) mice wcrc inoculated intraccrebraiiy (ie.) on
the light sidc of the bram as prcviously descrihcd (24) and using a 20-11.1 volume of
YF 17D or chimeric YF/WN cOnstructs for ncurovirulencc tcsting (titers of inocu
lated viruses are indicated in the Resuits section and in the tabies presentcd).
Mice were observed daily for 21 days foHowing inoculation to determine
survival ratio and svcrage survival time (AST) after virus chalienge.
Neutralizing antibody titers were detcrmincd by a constant virus-scrum dilu
tion 50% plaque rcduction neutralization assay test (PRNT
) in Vero ceils, as
5
previousiy described (24). An equal volume (0.1 mi) of virus suspension contain
ing 700 PFUImI and serial twofold dilutions of heat-inactivated scrum were
incubated ovcmight at 4°C, and the serum-virus mixture was inoculatcd orsto
Vero ccli monolayers grown in 12-well piates. An overiay of methyicellulose in
minimal cssentiai medium was added beforc incubation of the cultures at 37°C
for 3 to 4 daya prior to fixation and crystal violet staining for plaque count dc
termination. The endpoint neutralization titer was the highest dilution of serum
that reduced pisques by 50% compared to a mouse hyperimmune serum control.
Nonhuman primate studies. Neurovirulence tests in rhesus macaques were
performed according to World Hcalth Organization (WHO) guidelines for testing YF vaccine (36) and as described previousiy for safety tests of ChimeriVax-JE
vaccine (24). Animais were inoculated with specific virus candidates by inocula
tion of the frontal lobe of the bram (see Tahie 7). Blood sampies were obtaincd
daily for the first 10 days foliowing inocuiation, and serum viremia was measured
by plaquc assay on Vero celis. Animals werc obscrvcd daily for clinical signs of
encephalitis and associated symptoms such as fcvcr or tremors. Aninsals wcre
cuthanizcd 30 days after infection, and the bruin and spinal cord tissues were
removcd for histopathology. Slides wcre prepared from tissucs of the frontal and
temporal cortcx, hasai ganglia/thalamus (two leveis), midbrain, pons, cerebeilum
(two levels of the nuciei and cortex), meduila ohiongata, and six icvcis of cach of
ccrvicsi and lumbar eniargements of the spinal cord. Sections werc staincd with
gaiiocyanine. Histological lesions were snalyzed and scored for pathology rela
tive to that of the YF 17D according to the criteria for evaluation of neuroviru
lence in rhesus monkeys proposed by the current WHO requirements. Mean
lesion scores for individual monkeys were caiculated for “target” (suhstantia
nigra) and “discriminator” (basal ganglia/thalamus and the spinal cord) areas
individually and for the target and discriminator areas combined.
A second neurovirulence test was performed with cynomolgus monkeys and
inoculation with the YF/WN,R vaccine candidatc (ChimeriVsx-WN
) produc
02
tion virus secd (P4). The study was conducted according to good iaboratory
practices (OLP) standards (14a). Eicven monkeys were inoculated ie. with ‘/F/
WN)NR production virus seed (P4), 11 positive control monkeys received YF
VAX, and 5 negstive control monkcys rcceivcd dilucnt. The monkeys were eval
uatcd for changes in clinical signs (twice daily), body wcight (weekly), and food con
sumption (daily). Clinical signs were assigned scores according to a chnical scoring
system, baard on the WHO requirements for YF vaccine (36). Blood samples
were coliectcd premnocuiation on day 1 and on days 3, 5, 7, 15, and 31 for clinical
pathology analysis (serum chcmistiy and hematology paramcters). Additional blood
samples were coliected preinoculation on day 1 and on days 2 to 11 for viremia
analysis, and on days 1 (predose) and 31 for antiviral antibody titer analyses.
To determine immunogenicity, rhesus monkeys were inoculated by the sub
cutancous (s.c.) route with a single 0.5-mi dose containing —4 10810 PFU of
chimeric vaccine. Control animais received undiiuted YF-VAX contamning 4.49
iogiü PFU in a 0.25-mi volume. Each vaccine dose was back titrated following
immunization. Serum viremia was measured daily by piaque assay through day 10
after vaccination. Neutralizing antihody ieveis were measured hy PRNT
50 on days
14, 30, and 63 after vaccination. Animais were challenged 64 days after vaccination
by ie. inoculation of 125 pi contairilng 2.4 x 19’ PFIJ of wild-type WN NY99
suspendcd in M199 with HEPES huffer (Invitrogen) and 10% sorbitol (Sigma).
Monkcys were observcd for viremia, clinical ilinesa, and antibody rcsponse;se
vcreiy ili animais werc cuthanized. The i.c. challenge model closely followed the
model estsbiisshed during the dcvelopmcnt of ChirnetiVax-JE vaccinc (24, 26).
Genetic stabiiity (in vivo and in vitro passage) and sequencing. The chimeric
YF/WN virus containing unenodifled, wild.typc WN virus prME sequence (des
ignated ChimeriVax-WN
) was passcd six times in Vero E6 ceils foliowed hy
01
six passages in suckiing mice by the ie. route. The chimeric YF/WN virus
0
D
0
9)
0
0
-t,
-‘
0
3
t3
Ci)
3
b
-‘
CD
0
Ci,
0
1’3
0
(/2
H
0
0
1
0
0
0
1
m
0
0
1
fl1
VOL. 78, 2004
ChimeriVax-WEST NILE VACCINE PRECLINICAL EVALUATION
TABLE 2. Neuroinvasiveness of ChimeriVax-WN
, relative to
5
YF 17D based 011 dose response in 1CR mice’
Back titratjon dose
, PFU)
1
(log,
% Mortalitv
(no. dead/no. teted)d
01 (P2)
ChimeriVax-WN
5
0.89
2.23
3.24
4.06
5.45
6.51
0 (0/5)
0 (0/5)
0 (0/5)
0 (0/5)
0 (0/5)
0 (0/5)
YF17D (ATCC)
2.78
4.48
0 (0/3)
0 (0/3)
NA’
0 (0/3)
Test article
.
Negative control
.
•P•
1
Harlan-Sprague, 1CR strain (3 to 4-week-old female mice).
P2 indicates a sccond-generation passage virus on Vero celis. West Nile virus
strains are typically neuroinvasive after i.p. inoculation as shown by others (5).
NA. not applicahle.
d
AST was not deterrnined.
“
containing three mutations introdueed by site-directed mutagenesis (designated
,) al the P2 level (PMS) and P3 level (RMS) wcre passed 12
0
ChimeriVax-WN
and 10 timcs, respectively, in serum-free, protcin-free SF-Vero cdl substrate. All
in vitro virus passages were performed with an initial MOl of 0.01 PFU/cell
followed by harvest of the virus on the third day after infection. Passages in vivo
were performed by initial ie. inoculation of iO PFU; bram tissue from 1CR mice
(Taconie) was harvested 3 days after inoculation and homogenized, and the
clarified homogenate was used for passage to a new group of miee. Virus liters
al cach passage were determined hy plaque assay. Neurovirulence of the pas
saged viruses was determined hy ie. inoculation of adult or suckling mice (sce
Tables 13 and 14). Sequencing of vind RNA was perforrned with Superscript 11
reverse transcriptase and XL-PCR; products were purilied by QIAGEN gel
extraction (QIAGEN, Valencia. Calif.). Sequeneing reactions svere prepared and
analyzed using the standard Beckman CEQ 2000XL protocol and equipment
(Beckman-Coulter). For virus passages, at least two independent sequencing
reactions were executed per RT-PCR product strand sequenced; sense and
antisense strands wcrc sequenced each time. Mutation acccptance criteria
needed a positive identity in al Icast threc of four sequencing reactions analyzcd;
in addition, two indcpendent operators rcad sequenee chromatographs.
RESULTS
Virulence phenotype of chimeric YF/WN containing wildtype WN prME genes relative to YF 17D (YF-VAX). The
initial WN virus chimera encoded the envelope and premem
brane protein genes of the WN NY99 wild-type strain (desig
nated ChirneriVax-WN
). This chimeric virus did not cause
01
encephalitis after i.p. inoculation at doses of 1W’ PFU in 3- to
4-week-old adult 1CR mice (Table 2). Encephalitis was as
sessed by daily observations for illness, paralysis, and death.
, resembies YF 17D vaccine (33) in being
0
ChimeriVax-WN
nonneuroinvasive in adult mice. In contrast, the WN NY99
wild-type virus was lethal for mice when inoculated by the i.p.
route with as few as 1 to 4 PFU (5; unpublished results).
01 retained the ability to cause lethal en
ChimeriVax-WN
cephalitis after i.c. inoculation, a property consistent with
that of YF 17D virus (10). We estimated the i.c. 50% lethal
dose (LD
) of ChimeriVax-WN
50
01 to be between iO and io
PFU. The neurovirulence phenotype of ChimeriVax-WN
01 is
lower than that of YF 17D virus, for which the i.c. LD
50 is
between 10’ and 102 PFU (Table 3).
Evaluation of the multisite mutagenesis for attenuation.
Amino acids in the envelope protein previously established
as genetic determinates of virulence for ChimeriVax-JE
were changed to reduce the virulence of YF/WN chimeras.
12499
The strategy for this mutagenesis approach was to design a safe
attenuated WN vaccine; this strategy was first discussed in an
earlier publication (4). Briefly, the selection of specific amino
acid residues for mutagenesis was defined by previous studies
of the attenuating mutations in a vaccine strain of JE virus
(SA14-14-2) (3). Since the wild-type JE and WN viral E gene
sequences are identical at the residues implicated in attenua
tion of JE (SA14-14-2) vaccine, with one exception at residue
176, we postulated that introduction of mutations at the ma
jority of these sites into wild-type WN virus prME genes would
result in a similar attenuation of the WN phenotype. A.mino
acid residues mapping to the wild-type WN envelope (E) gene
positions 107, 138, 176, and 280 were all mutated in a single
construct to encode amino acid residues F, K, V, and M, respec
tively. The new chimeric virus was identified as YFfWN,KvM.
Chimeras were constructed in which each amino acid residue in
the FKVM group was individually mutated to produce single-site
mutants and to assess their individual roles in neurovirulence
(Table 4). The dissection of the FKVM group into single site
mutations identified only residue 107 as reducing virulence
significantly (0% mortality in three mouse neurovirulence tests
presented). Residue 280 followed with 0% mortality after a io
viral dose; however, inconsistency of this attenuated phenotype
(i.e., mortality ratios of 40 to 89%) was observed in the lower
viral-dose groups tested. A mutation at residue 138 resulted in
minimal reduction of virulence (—60% mortality), while a mu
tation at residue 176 showed no impact. The neurovirulence of
the multisite YF/WNFKvM construct resulted in 0 to 20% mor
tality. In later studies, amino acid residues 316 and 440 were
mutated to V and R, respectively, based on previous data in
dicating mutations in the E protein which mapped to these
regions thought to function in the biology of the E protein
third domain (1, 32). Changes in neurovirulence of these
mutants with respect to parental ChimeriVax-WN
111 were
evaluated in the mouse model as for the previous groups
above (Table 5). A single mutation at residue 316 resulted in a
greater attenuation (—30% mortality) than residue 440 but not
0
:3
0
0)
0
(0
0
1
0
3
Ci)
0
1
ca
—
0
:3
>
-D
-t
cri
c
0
(0
-1
0
w
()
TABLE 3. Neurovirulence of ChimeriVax-WN
01 relative to
YF 17D based 011 dose response in 1CR mice”
Test article ie.
01 (P2)
ChimeriVax-WN
YFI7D (ATCC)
Negative control
Back
titration dose
0 PFU)
(log,
5
% Mortality
(no. dead/no.
tested)
0
A T
(day’s)
20(1/5)
0 (0/5)
20(1/5)
4.06
60 (3/5)
5.45
20 (1/5)
9
9
0
0
0.9
0.98
2.78
20 (1/5)
60(3/5)
100(5/5)
100 (5/5)
100 (5/5)
9
10.3
9.2
8.2
8
NA’
0
1
fl.1
—2
—0.30
0.89
2.23
3.24
0 (0/5)
0 (0/5)
11
0
0
1
10
0 (0/3)
Harlan-Sprague, 1CR strain (3 to 4-week-old female mice).
Actual dose dclivered ie. assumed to he 20 11.1 for the back titration caleu
lations shown.
‘NA, not applicahle.
m
12500
J. ViIs.oL.
ARROYO ET AL.
01 (YF/WN)
TABLE 4. Neurovirulence of ChimeriVax-WN
site-directed mutagenesis variants at E protein residues 107—sF,
138—*K, 176—W, 280—s.M, tested by ic. inoculation in adult mice”
Test article
(Vero passage)
5
Target dose
(logj, PFU)
Back
% Mortalitv
titration done (no. dcadno.
tested)
PFU)
(log
AST
(days)
8.60
9
4
5
4.87
6.09
100 (5/5)
60 (3/5)
4
4
5
4.22
4.42
4.99
0 (0/5)
0(0/8)
0(0/5)
4
4
5
4.26
4.41
5.48
60 (3/5)
63 (5/8)
60 (3/5)
10.33
11.40
9.33
4
5
4.42
554
80 (4/5)
80(4/5)
12.50
11
M (P3)
50
YF/WN
4
4
5
4.14
4.55
5.14
40 (2/5)
89 (7/8)
0(0/5)
9
11.86
133 (P2)
F
107
YF/
KM
4
5
3.70
4.81
0 (0/5)
0(0/5)
F
107
YF/
1
K
VastM
35 (P3)
76
4
5
4.13
5.10
0 (0/5)
20 (1/5)
7
YF-VAX
3
2.77
100 (5/5)
9
WN NY99
4
3.90
100 (5/5)
5
, (P3)
0
ChimeriVax-WN
F (P2)
107
YFIWN
K (P3)
135
YF/WN
V (P3)
176
YF/WN
mutagenesis variants at E protein residues 107—F,
316—sV, and 440—.R tested in adult nice”
Back
Test article ie.
(Vero passage)
titration done
(log PF1.J)
% Mortality
(no. dead/no.
tested)
AST
(days)
01 (P3)
ChirncriVax-WN
4.11
4.74
4.83
83 (10/12)
60 (3/5)
100 (8/8)
9.20
10.33
10.63
V (P3)
115
YF/WN
4.09
4.67
4.57
25 (3/12)
38 (3/8)
38 (9/24)
12.33
10.67
11.22
R (P3)
441
YF/WN
4.17
4.60
4.35
83 (10/12)
38 (3/8)
56 (14/25)
9.22
10.33
11.21
3.90
4.12
3.71
17 (2/12)
40(2/5)
36 (9/25)
16.5
13
12
3.72
5.54
0 (0/12)
0 (0/12)
VR (P3)
315
YFfWN
4 (P4)
R
1
F
107
YFfWN
4
V
16
as significant as residue 107. The single mutation at residue 440
resulted in a greater level of attenuation over those at residues
138 and 176, but only in two of the three independent tests
performed (i.e., —40% niortality observed with a mutation at
residue 440). In summary, neurovirulence of the YF/WN chi
meras in which modified amino acids were inserted in the E
protein at residues 107, 316, and 440 were the most important
contributors to neurovirutence. Based on this information, a
multisite V.
316 construct was selected as our
F
107
YF/WN
R
0
vaccine candidate (ChimeriVax-WN
).
02
Neurovirulence studies in mice and in rhesus and cynomolgus
macaques. Neurovirulence of viruses with single or multisite mu
tations in the YF/WN virus E gene was measured in 21-day-old
mice inoculated by the i.c. route with doses between iO and i0
PFU. This assessment identified only residues 107 and 280 (Ta
ble 4) and the combination of residues 316 and 440 (Table 5)
as the dominant attenuating mutations as measured by mor
tality and AST. The chimera selected as our vaccine candidate
had mutations F, V, and R at residues 107, 316, and 440, respec
tively, and was avirulent for the adult mouse (Table 5). However,
this virus was neurovirulent for a 2-day-old suckling mouse (data
not shown). Because mice become resistant to flavivirus infection
in an age-dependent manner, the suckiing mouse is the most
sensitive host for determining subtle differences in neuroviru
, virus in suck
0
lence. Preliminaty studies with ChimeriVax-WN
ling nice of various ages showed that nice 8 days of age were able
to discriminate differences in neurovirulence, whereas younger
mice were too susceptible to differentiate the attenuation phe
02 vaccine candidate.
notype of the ChimeriVax-WN
A GLP study was undertaken to characterize the neuroviru
c
0
D
0
0)
0
CD
0
-
1
0
3
t3
Taconic, 1CR strain, 3 to 4-week-old female mice. Results of independent
experiments are shown.
Taconic, 1CR strain (3 to 4-week-old feniale niice).
P2 and P3 indicate second and third generation virus passage on Vero ceils,
respectively.
C,
01 site-directed
TABLE 5. Neurovji-ijlence of ChimeriVax-WN
11)
3
lence of the good manufacturing practice (GMP) manufac
, production virus seed (P4) and a
0
tured ChimeriVax-WN
vaccine lot (P5) prepared for clinical trials. Four litters (32
mice) of 8-day-old suckling mice were inoculated by the i.c.
route with 20 ii containing iO, iO, or iO PFU of either
production virus seed (P4) or vaccine (P5) virus. Control ani
mais of the same age received either lO
or iO PFU of YF
3
VAX. Negative controis were inoculated with diluent. The
results are shown in Table 6. There were no differences across
dose groups in the mortality ratios, and therefore data from
dose groups for each test article were combined for statistical
analysis. There was no difference in the mortality ratio of
animals infected with P4 or P5. Both the production virus seed
(P4) and the vaccine (P5) were highly attenuated compared to
YF-VAX. The neurovirulence profile of the WN vaccine is
therefore similar to that of the ChimeriVax-JE vaccine, which
is currently in phase II clinicai trials (25).
In a pilot monkey neurovirulence study, the ChimeriVax
01 construct was compared to that of the YF 17D vaccine.
WN
Rhesus macaques were screened and found negative for flavi
virus antibodies by hemagglutination-inhibition (HI) test
(kindly performed by Robert Shope, University of Texas Med
ical Branch, Galveston, Tex.). Groups of three young adult
1
to
0
0
1
01
t,’)
0
<
(0
H
0
132
0
0
0
1
ni
0
0
1
,
0
TABLE 6. Comparative neurovirulence of the ChimeriVax-WN
) vaccine candidate (P5), production virus seed
1
,V
3
F
107
(YF/WN
R
1
(P4), and YF-VAX in 8-day-old suckling 1CR mice (GLP study)
Test article
.
Negative control
02 P4 production virus seed
ChimcriVax-WN
, P5 vaccinc lot 02K01
5
ChimcriVax-WN
YF-VAX
% Mortalitv
(no. dead/no. tested)
-
0
1
4
98
(0/32)
(1/96)
(4/96)
(63/64)
Statistical significance was determined by Fisher’s exact test (two sided).
02 P5 versus YF-VAX, and P = 03684 for
P < 0.0001 for ChimeriVax-WN
, P4 versus P5.
0
ChimeriVax-WN
fl1
VoL. 78, 2004
ChimeriVax-WEST NILE VACCINE PRECLINICAL EVALUATION
12501
TABLE 7. Pilot study with rhesus monkeys of neurovirulence of ChimeriVax-WN
01 relative to YF-VAX based
on neuropathological evaluation at 30 days post-i.c. inoculations
Test article
.
YF-VAX
Monkey
Se
Back ttration dose
111 PFU)
(1og
Som of areas
M
4.40
0.59
F
4.40
0.5
0
0.64
P417
0.43
0.28
N555
F
4.40
1.5
0.66
0.94
0.58 ± 0.13
0.60 ± 0.33
0.67
N525
D402
C358
M
M
F
±
0.76
5.07
1.0
0.58
0.72
4.99
5.06
0.5
0
0.48
0.42
0.48
0.28
Mean ± SD
‘
Discriminator area
G211
Mean ± SD
01
ChimeriVax-WN
Individual histopathological score
Target area
0.50 ± 0.50
0.49
±
0.08
0
0.49 ± 0.22
M, male; F, fcmalc.
0
0
0
CD
0
-4,
1
0
rhesus monkeys were inoculated by the i.c. route with 5 log
10
PFU of ChimeriVax-WN
01 or 4.4 1og
10 PFU of commercial YF
17D vaccine (YF-VAX) (Table 7). Monkeys inoculated with
the chimera had a mean peak viremia titer of 1.85 ± 0.9 1og
10
PFU/ml with a mean duration of 4.5 days. Monkeys inoculated
with YF-VAX had a similar viremia profile (mean peak vire
mia titer of 2.65 ± 0.1 log
11 PFU/ml and a mean duration of
4.5 days). Histological scores induced by ChimeriVax-WN
01
were lower than those of a higher dose of YF-VAX (Table 7).
Histological lesions in all six monkeys were mildly infiamma
tory, predominantly small perivascular infiltrates. The vast ma
jority of them were scored as grade 1 on a scale of 1 to 4. No
involvement of neurons was seen. The lesions were located
mostly in YF vaccine discriminator centers (the basal ganglia!
thalamus areas and both enlargements of the spinal cord).
Comparison of the two groups of monkeys for the severity and
distribution of lesions did not reveal any noticeable difference.
On a second neurovirulence study, cynomolgus mon
keys were inoculated with YF/WNFVR vaccine candidate
) production virus seed (P4). These ma
02
(ChimeriVax-WN
caques were screened and found negative for fiavivirus anti
bodies by HI test (kindly performed by Robert Shope). Eleven
monkeys were inoculated i.c. with 4.74 1og
10 PFU of YF/WN
FVR production virus seed (P4), 11 reference control monkeys
received 5.34 log
1 PFU of YF-VAX, and 5 negative control
monkeys received diluent. The monkeys were evaluated for
changes in clinical signs (twice daily), body weight (weekly),
and food consumption (daily). Clinical signs were assigned
scores according to a clinical scoring system based on the
WHO requirements for YF vaccine (36).
YF 17D vaccine virus was detected in the sera of 10 of 11
monkeys inoculated with YF-VAX. The mean peak viremia ±
standard deviation (SD) was 357 ± 579 PFU/ml, and the mean
number of viremic days was 2.45 ± 1.13. Monkeyviremia titers
were below the 500 and 100 YF-VAX mouse ie. LD
50 values,
which are the maximum acceptable titers for individual mon
key and group viremia titers (i.e., present in no more than 10%
of the monkeys). respectively, as established under the WHO
requirements for YF 17D vaccine.
ChimeriVax-WN vaccine virus was detected in the sera of 10
of 11 monkeys inoculated with ChirneriVax-WN
, vaccine pro
0
duction seed bank (P4). The duration of viremia was 1 to 5 days
(mean, 2.9 ± 1.38) with peak titers ranging from 180 to 6,400
PFU/ml. The nuniber of viremic days did not differ between
treatment groups (P = 0.4067; analysis of variance [AVOVAJ).
A higher proportion of monkeys (91%) was viremic on the first
3
-r
day after inoculation than that seen in the YF-VAX group
(27%). On days 2 to 3 after inoculation, the proportion of viremic
monkeys (82%) was the same as for YF-VAX. The mean peak
viremiawas 2,097 ± 1,845 PFU/ml. Although the mean peakvire
mia titers for ChimeriVax-WN
, production virus seed (P4)
0
were higher than that of the reference YF-VAX vaccine (P =
0.0073; ANOVA), individual monkey and group viremia titers
for ChimeriVax-WN vaccine remained within acceptable group
and individual monkey specifications, based upon WHO require
ments for YF 17D vaccine (36). The WHO specifications stip
ulate that no individual monkey will have a viremia exceeding
500 ie. adult mouse LD
/ml and that no more than 10% of the
50
animals will have a viremia exceeding 100 i.c. mouse LD
/ml.
50
We have determined that these limits correspond to 20,000 Vero
PFU/0.03 ml and 4,000 PFU/0.03 ml, respectively, in the case
of YF-VAX (an LD
50 for ChimeriVax-WN
, cannot be deter
0
mined). The monkey viremias observed following ChimeriVax
WN do not exceed the limits set for YF vaccine.
07
There were no abnormalities in hematology or clinical chem
istry values associated with treatment. A complete necropsy
was performed on day 31, and tissues were prepared for his
topathology. There were no ChimeriVax-WN
, production
0
seed (P4)-related histopathologic changes in kidney, heart,
liver, adrenal glands, or spleen.
(0
3
1
CD
—
0
t3
-4
Oi
12
0
-&
F.)
0
(1)
H
(-)
tIJ
0
-J
G)
0
T
m
0
0
TABLE 8. Summary of CNS histopathologic lesion scores in
cynomolgus monkeys inoculated by the i.c. route with ChimeriVax
, production virus seed (P4), YF-VAX, or negative control
0
WN
Mean z SD lesion scores
Treatment group
Negative control
,
11
ChimcriVax-WN
production virus
seed (P4)
YF-VAX
P-value’
Target
areas
5
11
0
0.12 ± 0.11
11
0.5 ± 0.22
0.000476
Discriminator
areas
Combined
score
0
0.13
±
0
0.13
0.54 ± 0.23
0.000357
0.13
±
0.09
0.52 ± 0.2
0.000122
The Kruskall-Wallis test was used for comparison of the ChimeriVax and
YF-VAX groups.
1
m
12502
J. Vn.oi.
ARROYO ET AL.
TABLE 9. Neutralizing antibody titers (PRNT
) and protective
50
high titers of neutralizing antibodies to the respective virus
with which they were inoculated (data not shown).
Immunogenicity and efficacy studies in mice and rhe
, and
0
sus monkeys. The immunogenicity of ChimeriVax-WN
, was evaluated in adult 1CR mice inoculated
0
ChimeriVax-WN
by the s.c. route. Serum neutralizing antibodies were measured
by PRNT
4 weeks after vaccination with a single dose, and titers
50
were expressed as the geometric mean titer (GMT) (Table 9).
, vaccine elicited antibody titers
0
In mice, ChimeriVax-WN
that were approximately lO-fold lower than those elicited by
01 virus, reflecting the greater attenuation of
ChimeriVax-WN
this virus. However, when mice were challenged i.p. with 1,000
50 of wild-type WN NY99, mice that had been immunized
LD
, or
0
with either ChimeriVax-WN
were protected in a dose
dependent manner. A vaccine dose of io PFU of ChimeriVax
02 protected all animals, whereas a dose of io PFU pro
WN
tected only 40% of the animals.
Young adult rhesus monkeys seronegative for WN neutral
izing antibodies were vaccinated by the s,c. route with three dif
ferent chimeric vaccines: (i) a chimera containing the E107 (L—*
F) single-site mutation (YF/WNF); (ii) a chimera containing two
mutations at E316 (A—.V) and E440 (K—>R) (YF/WNVR); and
(iii) ChimeriVax-WN
, containing all three mutations.
0
Viremia in the monkeys immunized with the different
ChimeriVax-WN viruses following s.c. inoculation was longer
relative to YF-VAX in some animals, although the levels de
tected at later time points were very low (Table 10). Viremias
in monkeys receiving the ChimeriVax-WN vaccines ranged
50 PFU/ml, with a mean duration of 3.5 to 5
from 1.0 to 2.3 log
days. The mean peak titers of the viremia in monkeys given
YF-VAX were approximately the same as those receiving the
WN vaccines. Among the ChimeriVax-WN vaccines, the vire
mia titers measured suggest an inverse relationship between
activity of ChimeriVax-WN candidate vaccines in
adult 1CR mice challenged by the i.p. route”
Wild-type
WN NY99
PRNTa,
% Survival
(no. live!
total)
Vaccine
s.c. dose
5 PFIJ)
(log,
GMT ± SD
(4 wk posts.c. vaccine)
01
ChimeriVax-WN
3.48
197 ± 93
3
100 (8/8)
ChimcriVax-WN,
2.64
5.01
20 ± 0
37±45
3
3
40 (4/10)
100(9/9)
NA”
0
3
0 (0/5)
Negative control
challenge
i.p. dose
10 PFU)
(1og
“ Micc wcrc challcngcd 4 weeks after s.c. vaccination (challengc titer was not
back titrated).
“NA, not applicable.
(
Histopathology of the bram and spinal cord was performed
according to the methods described by Levenbook et al. (20)
and incorporated into the WHO requirements for YF vaccine
(36). Central nervous system (CNS) lesions were noted in 11 of
,
0
11 and 10 of 11 of YF-VAX-treated and ChimeriVax-WN
vaccine-treated monkeys, respectively, and there were no CNS
lesions in the vehicle control monkeys. Infiammatory lesions
induced by both viruses in the meninges and the bram and
spinal cord matter were minimal to mild (grades 1 or 2) and
composed of scanty, mostly perivascular infiltrates of mononu
dear celis. There was no involvement of neurons in any of the
- or YF-VAX-treated monkeys. Summary
02
ChimeriVax-WN
data are presented in Table 8. ChimeriVax-WN production
virus seed (P4) was signiflcantly less neurovirulent (P < 0.05)
than the reference article, YF-VAX, in the target, discrimina
tor, and combined rnean lesion scores. All monkeys developed
0
ZI
0
0.)
0.
(t’
ci.
-t,
1
0
3
:3-
-a
0)
t,,
3
b
-‘
(0
0
0
1
(ii
0
-
0
(1)
-1
TABLE 10. Viremia in rhesus monkcys inoculated by the s.c. route with YF-VAX, ChimeriVax-WN virus constructs with single or
, vaccine candidate
0
double mutations, and the ChimeriVax-WN
Vaccine and
monkey
Viremia (1og
10 PFU/ml) at day postinoculation”:
Vaccine dose
(log, PFU)
1
2
3
4
5
YF-VAX
M017
BlOl
R286
T081
4.49
4.49
4.49
4.49
0
0
0
1.3
1.0
1.6
1.8
1.0
2.1
2.0
2.8
1.5
2.9
1.9
2.6
2.0
2.4
0
0
0
0
0
0
0
YF/WN,
F
07
N313
P367
TOS7
AESI
4.19
4.19
4.19
4.19
1.6
0
2.3
2.3
2.0
1.7
2.3
2.1
1.0
1.6
1.3
1.6
1.3
1.8
1.3
1.3
1.0
2.3
0
0
YF/WN
V,,R
315
R918
N577
M233
T757
4.0
4.0
4.0
4.0
0
1.0
0
0
2.0
1.9
0
0
1.5
1.5
1.0
0
1.7
1.0
1.0
1.6
YF/WN
2729
T445
T086
T491
3.92
3.92
3.92
3.92
1.0
1,0
1.0
0
0
1.6
0
1.5
0
1.5
1.3
0
1.0
0
1.3
1.0
“
6
Mean peak
titer ± SD
Meso duration
0
0
0
0
2.4 ± 0.5
3.5
0
1.3
0
0
0
0
0
0
2.2 ± 0.2
0
0
1.0
0
0
0
1.6
0
0
0
1.8
0
1.8 ± 0.2
3.5
1.0
0
0
0
0
0
0
1.0
1.0
1.0
0
0
1.4 ± 0.2
4.5
7
8
9
0
0
0
0
0
0
0
0
0
0
0
0
0
1.6
0
0
0
0
0
0
0
0
0
1.0
0
0
0
0
0
1.0
0
0
0
0
0
0
1.3
0
0
0
1.0
1.0
0
0
0
0
0
1.0
10
(days)
0
0
0
G)
0
m
0 PFU/ml is the assay tower limit.
No virus was detected in the assay at day 0 (preinoculation); 1.0 log,
5
C)
0
t
>
m
VoL. 78, 2004
ChimeriVax-WEST NILE VACCINE PRECLINICAL EVALUATION
TABLE 11. Reciprocal neutralizing antibody titers (PRNT
)
0
against ChimeriVax-WN virus, rhesus monkeys inoculated by the
s.c. route wiih YF-VAX or ChimeriVax-WN vaccine candidates
PRNT) on da?:
Vaccine and
monkey”
YF-VAX
M017
BIOl
R286
TOSi
Dose
10 PFU)
(1og
4.49
GMT
YF/WN
F
17
N313
4.19
P367
T087
Postimmunization
14
30
<320
<320
<320
640
>640
>640
>640
>640
380
640
160
<40
<40
>640
640
640
>640
57
640
453
4,305
4,305
<40
320
>640
>1,280
2,560
320
320
>640
>1,280
640
>1,280
2,560
>5,120
453
1,076
2,560
<40
AE81
GMT
Postchallenge
63
15
>5,120
NA’
2,560
1,280
640
NA
2,560 >10,240
2,153
3,620
31—34
NA
2,560
NA
5.120
3,620
>640
2,560
5,120
640
5,120
2,560
>640
2,560
1.280
160 >10,240 >20,480
3 6
YFfWN
440
V
R918
4.0
N577
M233
T757
<40 160—320e
<40 160—320e
<40
40
GMT
YF/WNR
J729
T445
T086
T491
GMT
40
3.92
135
<40
320
80
640
160 320—64(Y
80
320
80
381
1,280
80
160
>640
160
>5,120
640
1,280
2,560
>5,120
>5,120
>5,120
>5,120
190
1,280
5,120
For GMT, if an endpoint was not rcached, the assay limit titcr was used in the
caiculation (eg., >640 taken as 640 and <40 was taken as 40).
0 was calculated after subtraction of the PRNT from day 0 serum
PRNT
samples.
1 caiculation feit bctwccn the titers shown. The lower titer was used
PRNT.,
for the OMT caleuiation.
aNA not appilcabie.
the number of attenuating mutations in the chimera and the
peak titer of viremia (Table 10), but small sample size pre
cludes definitive characterization of these differences.
The immunogenicities of vaccine candidates with one, two,
or three attenuating mutations were similar (Table 11). Neu
tralizing antibody titers ranged from 40 to >640 depending
upon the vaccine. There were no significant differences in
neutralizing antibody response between treatment groups (Ta
bie 11). High titers of neutralizing antibodies (>100 PRNT
)
50
were present 30 and 63 days after vaccination. The observation
that monkeys developed neutralizing antibodies by day 14 in
dicates that ChimeriVax-WN
02 elicits rapid onset of protective
immunity.
Rhesus rnonkeys vaccinated with YF-VAX developed neu
tralizing antibodies against YF 17D with GMTs of 380 on day
14 postvaccination and 2,153 by day 63, which was 1 day before
challenge with the virulent WN NY99 virus.
Monkeys immunized with ChimeriVax-WN single, double or
triple mutants were uniformiy protected against lethal i.c. chal
12503
lenge with WN NY99 (Table 12). It is noteworthy that 50% of
the animals vaccinated with ChimeriVax-WN developed fever
after challenge, with an average duration of 5 days postchal
lenge, suggesting that they sustained subclinical infections. An
i.c. challenge with WN virus is extremely aggressive and is the
only route of challenge tested to induce WN virus disease in
naïve rhesus monkeys. It is likely that virus replication occurs
in bram tissue after i.c. inoculation and before a specific im
mune response in the bram can be recruited for clearance of
the virus. In the case of a human peripherally challenged by a
mosquito bite, preexisting immunity would rapidly neutralize
the virus and fever is unlikely to occur. However, none of the
ChimeriVax-WN-immunized animals developed detectable
viremia after challenge, none developed signs of iiiness (aside
from fever), and none died. Vaccinated animals showed an
increase in antibody levels postchallenge (Table 11), suggesting
that viral replication and antigenic stimulation occurred with
Out associated illness.
Postchallengeviremias (.102 to iO PFU/ml) were detected
in the control monkeys that had previously been immunized
with YF-VAX (Table 12). Two Out of four monkeys vaccinated
with YF-VAX (M017 and R286) developed a high fever and
signs of encephalitis: muscie tremors, anorexia, and spasticity.
These two animals were euthanized between days 9 and 11
after challenge. The other two YF-VAX-vaccinated animals
developed fever and survived ie. challenge with WN NY99
strain without any cLinical symptoms; this finding is attributed
to cross-protection across the two flaviviruses.
Two monkeys without any prior vaccination were also chal
0
0
:5
0
0)
0
0
-
1
0
3
1’
(1)
3
ö
1
to
0
:5
>
-t
01
0
TABLE 12. Viremia and clinical outcome in rhesus monkeys
immunized with ChimeriVax-WN or YF-VA.X and
chalenged 63 days later by the i.c. route with
5.38 1og
10 PFU of wild-type WN NY99 virus
Viremia by day post-i.c.
chalienge (Iog PFU/rnl)
Vaccine and
monkey
1
2
3
4
No. of monkeys with
outcome/total (%)
5
lllness
Neg control
K396
P500
2.0
2.0
3.1
3.1
2.6
2.6
2.4
2.2
0
1.0
2/2 (100)
YF-VAX
M017
BlOl
R286
T081
2.5
2.6
3.3
2.4
2.6
2.5
3.3
3.0
1.7
1.7
1.7
2.4
1.3
0
0
0.5
0
0
0
0
2/4 (50)
Death
2/2 (100)
t”)
(0
H
0
w
2
‘*,1
0
2)4 (50)
0
1
m
F
YF/WN 107
N313
P367
T087
AE8I
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0/4(0)
YF/WN R
1
11
V
16
R918
N577
M233
T757
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0/4(0)
0/4(0)
ChimeriVa.x-WN
,
0
J729
T445
T086
T491
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0/4 (0)
0/4 (0)
0/4(0)
0
0
1
m
12504
J. ViisoL.
ARROYO ET AL.
TABLE 13. Neurovirulence of YF/WNFvR RMS (P4 and Pil)”
and PMS (P2 and PlO)” in 2-day.old 1CR strain
mice relative to YF 17D”
Virus i.c.
Mutation
Back
titration dose
10 PFU)
(1og
Mortality
(no. deadJtotal)
AST
(days)
ChimeriVax.WN
2
RMS
P4
Pil
None
E336C—’S
1.73
2.08
80 (8/10)
60 (6/10)
14.5
13.67
ChinicriVax.WN
,
0
PMS
P2
PlO
None
E313G—’R
2.10
1.88
60(6/10)
70(7/10)
13
13.67
YF-VAX
NA’
1.90
100(10/10)
10.6
NA
0 (0/10)
Negative control
Viruscs were passed in a scrum-frec (SF-Vcro) stationary-ccll substrate.
b
Taconic, 1CR strain, mice.
‘NA, not applicable.
lenged with WN NY99 virus. The two challenge control ani
mais developed fever between days 5 and 9 postchallenge, with
slight tremors progressing to ataxia and spasticity between days
10 and 11, and were euthanized between days 10 and 12.
Genetic stability. In vitro and in vivo substrate-passage stud
01 or the YF/WN,R chimeric vac
ies with ChimeriVax-WN
) were conducted to deter
02
cme candidate (ChimeriVax-WN
mme genetic stability of the constructs when grown in
stationary ceil cultures and in bram tissue. After six in vitro
(‘
Vero E6 celi passages of the virus followed by six in vivo 1CR
, no muta
1
adult mouse bram passages of ChimeriVax-WN
tions were selected relative to the wild-type sequence of the
01 construct nor was
prM and E genes in the ChimeriVax-WN
there an increase in mouse neurovirulence (data not shown). A
heterozygous mutation in the E protein at position E336 re
sulting in a cysteine-to-serine change was identified following
10 in vitro passages of the YFIWNFvR virus in Vero E6 cells.
In a separate study, in vitro passage of YF/WN,R in SF-Vero
celis (manufacturing substrate) resulted in selection of a mu
tation at position E313 that changed the amino acid at that
position from glycine to arginine. Neurovirulence of these pas
saged viruses for the 2-day-old suekling mice (ii = 10) inocu
10 PFU dose of viruses inciuding
lated with a norninal 2-1og
E313 and E336 mutations showed no increase in virulence
relative to YF/\VNFvR PMS (Table 13). During all serial pas
sages of the virus in Vero ceils or bram tissue, no reversions
were detected at target E protein amino acid residues 107F,
316V, or 440R, the attenuation markers for the vaccine
candidate. Additionally, during scale-up manufacturing of the
., vaccine, no reversions at these critical res
0
ChimeriVax-\VN
idues were detected.
02 production vi
The GMP manufactured ChimeriVax-WN
rus seed (P4) was used for inoculation of large-scale Vero-SF
cultures grown on microcarrier beads in 100-liter bioreactors.
An additionai mutation (L—*P) occurred in the vaccine at
position 66 in the M protein. This mutation was associated with
production of slightly smaller plaque size. The vaccine lot (P5)
contained equal ratios of small and large plaques. Virus pop
ulations with and without the M66 mutation were isolated by
plaque purification and compared to the PMS (no detectable
mutations) and the vaccine lot in the suckling mouse model.
One litter (10 mice) of 8-day-old mice was inoculated by the i.c.
10 PFU of either large
containing 2, 3, or 4 1og
route with 20
plaque or small-plaque virus and observed for 21 days for signs
of illness and death. For comparative purposes, litters of mice
were inoculated with similar doses of the PMS (P2) and vac
cme lot (P5) viruses. Mice of the same age were also inoculated
() PFU of YF-VAX. Negative con trois were inocu
1
with 2 log
lated with diluent (Table 14). There were no differences in
mortality ratios across dose groups, and data were combined
for analysis. Since the mortality ratio across all treatment
groups differed (P < 0.0001), pairwise comparisons were per
formed. The M66 mutation had no effect on mouse neuroviru
lence.
0
0
0
0
0
-1,
1
0
3
Z3
t,,
3
b
1
CQ
0
Z5
1
ci,
0
DISCUSSION
-
The original YF/WN chimeric virus constructed by insertion
of the prME genes from a wild-type WN virus strain was
attenuated with respect to the parental YF 17D virus vector,
but retained a degree of neurovirulence for adult mice. To
develop a vaccine candidate with a wider margin of safety, we
selectively introduced mutations in the donor WN virus. Mu
tations introduced into the E protein of the WN donor virus
utilized a strategy based on the previous construction of
ChirneriVax-JE vaccine, which contained donor prME genes
from an attenuated vaccine strain of JE (SA14-14-2 virus) (3,
4, 28). The SA14-14-2 virus contains mutations at six amino
acid residues (E107, E138, E176, E279, E315, and E439) that
(1W)
-1
0
ci
0
1
0
0
0
1
m
0
0
TABLE 14. Neurovirulence of small- and large-plaque viruses isolated from ChimeriVax-WN(), P5 vaccine in
8-day-old 1CR mice inoculated ie.”
1
P value for
:
5
Test article
Mutation
% Mortality
(no. dead/no. lested)
Test article vs
Negative control
Sham (negative control)
PMS (P2)
Vaccine lot (P5; large and small plaque)
Large plaque
Small plaque
YF-VAX
None
E313G—uR, M66L—’P
E313G—’R
E3I3G—sR, M66L—sP
0
13
23
3
13
(0/10)
(4/30)
(7/30)
(1/30)
(4/30)
100 (10/10)
0.5558
0.1612
1.000
0.5558
YF-VAX
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
m
Large plaque vs
small plaque
0.3533
<0.0001
15
7 Additional mutations appearing during Vero
4
5
,
3
F
4
V
, Viruses used in the cvaluation containcd the sitc.directcd attcnuating mutations ,,R.
0
All ChimcriVax-WN
cdl passage are shown in the table.
Fisher’a exact test (two sided).
VOL
78, 2004
ChimeriVax-WEST NILE VACCINE PRECLINICAL EVALUATION
play a role in neurovirulence (3). The WN and JE wild-type
gene sequences are conserved at most of these residues (except
176), suggesting that mutations introduced at these sites in WN
virus could have the same attenuating effect as they did in the
case of JE SA14-14-2. As predicted, we found that mutagenesis
of the WN E residues E107, E280 (corresponding to E279 in
JE virus), and E316 (corresponding to E315 in JE virus) caused
attenuation of the YF/WN virus chimera. Surprisingly, while
an E138 mutation, E—K, was associated with a marked atten
uation of JE virus (3, 34), a corresponding mutation in the WN
gene did not reduce the neurovirulence of the YF/WN virus to
the expected 0% mortality by the mouse neurovirulence test.
Mutation of the E protein at E440 (corresponding to E439 in
JE virus) from K—+R, a conservative residue change, also re
duced neurovirulence for mice. A construct with the three
mutations of F, V, and R at positions E107, E316, and E440,
respectively, was designated ChimeriVax-WN
02 and was se
lected as the candidate for manufacture of the vaccine for
clinical studies. ChimeriVax-WN
, was not neuroinvasive corn
0
pared to WN NY99 virus and had reduced neurovirulence
compared to YF 17D vaccine virus. Attenuation of this virus
was conferred by the mutation at E107, which maps to the
fusion peptide in the second domain as predicted in the crystal
structure of the E protein (1, 12, 32). This amino acid is
thought to reduce virulence by altering the function of the
fusion peptide in the natural cycle of the virus replication. The
additional ChimeriVax-WN
, mutations at positions E316 and
0
E440 map in domain III on the crystal structure of the E
protein. Residue E316 is thought to be involved in binding of
tick-borne encephalitis virus to the virus receptor on the ceil
plasma membrane (1, 32) and thus may play a role in WNvirus
celI ently. Residue E440 is in the transmembrane region of the
E protein and is believed to be involved in anchoring the E
protein during its translation in the endoplasmic reticulum;
hence, a mutation at E440 may be altering the natural associ
ation of the E protein with prM (2). The K-to-M mutation at
position E280 that attenuated neurovirulence for mice was not
included in the final vaccine because It appeared unstable,
similar to the corresponding residue in JE virus E protein
sequence (i.e., E279) shown to be unstable during in vitro
passage. A reversion to K at position 279 in the JE virus E
protein occurred after less than five passages of the virus in
MRC-5 ceils (22). Mutation of residue E176 from Y in the WN
virus sequence to either V or T, as seen in JE strains, did not
suggest a significant change in neurovirulence; therefore, po
sition E176 was not changed in the final vaccine candidate
sequence (unpublished results). This observation contrasts to
the previously published results linking a mutation from 1 to V
at position E176 in the JE virus envelope protein to neuro
virulence (3, 28). Other approaches to flavivirus chimeras em
ployed an attenuated dengue virus genome backbone to pro
duce chimeric dengue virus vaccine candidates against the four
major serotypes (15); similarly, a dengue virus has been used to
deliver the prM and E genes of WN virus, producing an atten
uated vaccine candidate shown protective in a nonhuman pri
mate model (30). This dengue/WN virus chimeric construct
was attenuated by virtue of the chimeric nature and as a result
of a 30-nucleotide deletion in the 3’ end noncoding region
(untranslated region) of the virus genome.
Safety of ChimeriVax WN
02 (YF/WNFVR) is characterized
12505
by three features: (i) loss of neuroinvasion relative to wild-type
WN virus; (ii) introduction of three site-directed mutations in
two E protein domains, each independently associated with
attenuation; and (iii) conservation of the FVR mutations after
in vitro passage in manufacturing-related substrates.
The safety of ChimeriVax-WN
02 was evaluated in a sensitive
8-day-old suckling mouse model and in rhesus monkeys and in
cynomolgous macaques inoculated by the i.c. route. In all hostvirus pairings, the chimeric virus proved to be significantly less
neurovirulent than the licensed YF-VAX vaccine. The monkey
safety test was performed as prescribed by current regulations
applicable to YF vaccines (36) and showed that the vaccine was
significantly less virulent than YF-VAX. The nonhuman pri
mate model has been previously used to assess the safety of
other chimeric vaccines against JE and dengue virus (6, 18, 24).
After s.c. inoculation of rhesus monkeys, viremias were more
erratic and of longer duration in animals immunized with the
ChimeriVax-WN vaccines than in animals given YF-VAX (Ta
bIe 10). The mean peak titer viremia for YF-VAX-vaccinated
monkeys was —1 log higher than that for the ChimeriVax
02 (triple mutant) vaccine candidate. The longer viremia
WN
observed after immunization with the chimeric viruses suggests
that the viruses replicate in different tissues had different re
ticuloendothelial clearance rates from the parental YF 17D
virus or had different kinetics of immune response. We are
currently studying the sites of replication of ChimeriVax-WN
,
0
and YF-VAX in tissues of cynomolgus macaques and will
report results in a future publication. In addition, future clin
ical trials will assess the magnitude and duration of viremia
following ChimeriVax-WN
, and YF-VAX and establish cor0
relations between viremia and adverse events. The low titer of
the viremia observed in rhesus monkeys after s.c. vaccination
with the chimeric vaccine candidates suggests that ChimeriVax
WN(), vaccine has an acceptable phenotype for trials in hu
mans.
The triply mutated virus (ChimeriVax-WN
) vaccine ap
02
peared to be less immunogenic than the wild-type chimera in
mice, but performed satisfactorily in nonhuman primates.
02 vaccine rapidly elicited a neutralizing anti
ChimeriVax-WN
body response in all rhesus monkeys and provided solid pro
tection against an aggressive i.c. challenge with 5 log
10 PFU of
WN NY99 virus.
A partially protective immune response was observed in two
of the four rhesus monkeys immunized with YF 17D and
subsequently challenged with wild-type WN virus. Previous
observations by others have shown the cross-protective effect
of prior exposure to phylogenetically related flaviviruses and
concluded that potential for protective cross-reactivity is un
likely to prevent infection and only likely to prevent disease
(16). Similarly, we observed that prior YF immunization of
monkeys did not prevent infection (viremia) after WN virus
challenge, but may have provided an element of protection
against death. It should be pointed out that the interval be
tween YF immunization and challenge was relatively brief and
that cross-protection between heterologous flaviviruses often
diminishes over time, probably due to affinity maturation of the
antibody response and waning of T-cell immunity. It is highly
unlikely that YF immunity would provide reliable cross-pro
tection of humans and therefore a specific, homologous (WN)
vaccination strategy must be pursued. This observation is sim
0
0
0
0
0
CD
0
-,
0
3
-‘-
(1)
3
b
-‘
CD
0
>
0
1
Cii
r\J
c
0
(-t)
H
0
UJ
1;
25O6
Q
Ç.
ARROYO ET AL.
ilar to a previous report that hamsters vaccinated with YF 17D
were somewhat cross-protected against WN virus challenge
induced disease (35). However, in the rhesus model, only those
animals immunized with the WN vaccines and subsequently
challenged with wild-type WN virus i.c. did not show postchal
lenge WN virus viremia. All of the surviving animals vaccinated
with ChimeriVax-WN displayed increases in WN neutralizing
antibodies after i.c. challenge indicating that the challenge
virus had replicated in bram tissue (without causing illness) or
that the challenge inoculum, which was quite large, provided
sufficient antigen for stimulation of B ceils. The experimental
design did not allow a proper test of whether the preexisting
immunity would have been “sterilizing” if the challenge moe
ulum had been delivered by a natural (parenteral) route in
stead of i.c. Sterile immunity could be tested by measuring the
immune response to nonstructural proteins of the WN chal
lenge virus. Others have reported that experimental WN vac
cines elicit sterilizing immunity against parenteral challenge
(11).
Since the first experimental vaccine construct with wild-type
prME sequence was less neurotropic than commercial YF vac
cme, the chimeric vaccine with three attenuating mutations has
a wide margin of safety. To maintain this ultra-attenuated
phenotype, the vaccine candidate must retain attenuating mu
tations FVR introduced to retain the level of attenuation re
quired. Because of the quasispecies nature of RNA viruses,
variations in the sequence are to be expected during vaccine
manufacture. Therefore, quality of the product is carefully
monitored during manufacture by tests for genotypic and phe
notypic stabiity. Safety is ensured by the demonstration of
conservation of the amino acid residues identified to play a
role in attenuation (shown by direct sequencing release tests);
for ChimeriVax-WN
, the required conserved residues are E
02
protein 107F, 316V, and 440R. In addition, the attenuated
virulence phenotype of the vaccine is tested by infant mouse
neurovirulence test performed on seed viruses and each vac
cme batch. Currently, the product specifications for sequence
data have been expanded to inciude full genomic sequencing of
cach vaccine hatch rather than confirmation only of the point
mutations at residues 107, 316, and 440. When ChimeriVax
02 virus was passed in Vero cells, with al least twice the
WN
number of passages required for manufacture of the vaccine,
the FVR mutations were maintained. Passage of the vaccine
candidate, in vitro or in vivo, selected mutations in the vicinity
of residue E316 (at E313 and al E336) without compromising
the neurovirulence phenotype of ChimeriVax-WN
02 and sup
porting our mutagenesis approach to ensure vaccine safety
When the vaccine was scaled up for manufacture of clinical
material in 100-liter bioreactors, a mutation at M66 was de
tected. This mutation also did not affect neurovirulence or
immunogenicity of the vaccine.
J. Vioi.
Rosa and Robert B. Tesh; and from Acambis, mc., Rich Weitzin,
Zheng-Xi Zhang, Jian Liu, and Rick Nichols. Thanks go to Denise
Goens for critical review of the manuscript.
REFERENCES
1. Allison, S. L., J. Schalich, K. Stiasny, C. W. Mandi, and F. X. Irleinz. 2001.
Mutatiortal evidence for an intemal fusion peptide in flavivirus errvelopc
protein E. J. Virol. 75:4268-4275.
2. Allison, S. L., K Stiasny, K. Stadier, C. W. Mandl, and F. X. Heinz. 1999.
Mapping of functional elements in the stem-anchor region of tick-bome
encephalitis virus envelope protein E. J. Virol. 73:5605—5612.
3. Arroyo, J., F. Guirakhoo, S. Fenner, Z.-X. Zhang, T. P. Monath, and T. J.
Chambers. 2001. Molecular basis for attenuation of neurovirulence of a
yellow fever virus/Japanese encephalitis virus chimera vaccine (ChimeriVax
JE). J. Virol. 75:934—942.
4. Arroyo, J., C. A. Miller, J. Catalan, and T. P. Monath. 2001. YelIow fever
vector live-virus vaccines: West Nile virus vaccine development. Trends Mol.
Mcd. 7:350—354.
5. Beasley, D. W., L Li, M. T. Suderman, and A. D. Barrett. 2002. Mousc
neuroinvasive phenotypc of West Nile virus strains varies depending upon
virus gcnotypc. Virology 296:17—23.
6. Butrapet, S., J. Rabableri, S. Angsubhakorn, W. Wiriyarat, C. Huang, R.
Kinney, S. Punyim, and N. ilhamarapravati. 2002. Chimeric dcngue type
2/type 1 viruses induce immune responses in cynomolgus monkcys. South
east Asian J. Trop. Mcd. PubI. Health 33:589—599.
0
0.
CD
0
-h
-,
0
3
:3-
7. Centers for Disease Control and Preventiun. 1999. Outbreak of West NiIc
like viral encephalitis—New York, 1999. Morb. Mortal. Wkly. Rep. 48:845—
849.
8. Centers for Disease Control and Prevention 2003. West Nile virus activity—
United States, October 16—22, 2003. 52:1022—1023.
9. Chambers, T. J., A. Nestorowicz, P. W. Mason, and C. M. Rice. 1999. Yellow
fever/Japanese encephalitis chimeric viruses: construction and biological
properties. J. Virol. 73:3095—3101.
10. Chambers, T. J., and M. Nickeils. 2001. Neuroadapted yellow fever virus
t,
17D: genetic and biological characterization of a highly mouse-neurovirulent
virus and its infectious molccular done. J. Virol. 75:10912—10922.
Ilavis, B. S., G.J. J. Chang, B. Cropp, J. T. Roehrig, D. A. Martin, C. J.
Mitchell, R. Bowen, and M. L. Bunning. 2001. West Nilc virus recombinant
DNA vaccine protccts mousc and horse from virus challcngc and exprcsscs
in vitro a noninfcctious recombinant antigen that can lie used in enzyme
linked immunosorbent assays. J. Virol. 75:4040—4047.
Despres, P., M. P. Frenkiel, and V. Deubel. 1993. Differences bctwecn cdl
membrane fusion activities of two dengue type-l isolates reflect modifica
tions of viral strudture. Virology 196:209—2 19.
Dupuis, A. P., P. P. Marra, and L. D. Kramer. 2003. Serologic cvidencc of
West Nile virus transmission, Jamaica, West lndies. Emerg. Infect. Dis.
9:860—863.
Fernandez-Salas, 1., J. F. ContrerasCordero, B. J. Blitvich, J. 1. Gonzalez
Rojas, A. Cavazos-Alvarez, N. L. Marlenee, A. Elizondo.Quiroga, M. A.
Lorono-Pino, D. J. Gubler, B. C. Cropp, C. H. Calisher, and B. J. Beaty.
0
D
11.
12.
13.
14.
2003. Serologic evidence of West Nile virus infection in birds, Tamaulipas
State, Mexico. Vector Bome Zoonotic Dis. 3:209—213.
14a.Food and Drug Administration. 2001. 21 CFR, part 58. Good laboratory
practice for nonclinical laboratory studies. Food and Drug Administration,
Rockvillc, Md.
15. Huang, C. Y.-H., S. Butrapet, K. R. Tsuchiya, N. Bhamarapravati, D. J.
Gubler, and R. M. Kinney. 2003. Dcngue 2 PDK-53 virus as a chimeric
carrier for tetravalent dengue vaccine development. 3. Virol. 77:11436—
11447.
16. Kanesa-Thasan, N., J. R. Pulnak, JA. Mangiafico, J. E. Saluzzo, and G. V.
Ludwig. 2002. Short report: absence of protective ncutralizng antibodies to
17.
18.
19.
ACKNOWLEDGMENTS
This work was funded by a NIAID ROl grant A148297 and NIH
grant 5P51-RR00164-41.
We would like to aeknowledge contributions from the following:
independent contributor Inessa Levenbook; from University of Mas
0
20.
sachusetts Medical School, Worcester, Sharone Green, Francis Ennis,
and John Cruz; from the Centers for Disease Control and Prevention,
John Roehrig; from Sierra Division, Charles River Laboratories, Ken
21.
Draper; from University of Texas, Galveston, Amelia P. Travassos da
22.
West Nile virus in subjects follosving vaccination with Japanesc encephalitis
or dengue vaccines. Am. J. Trop. Mcd. Hyg. 66:1 15—1 16.
Komar, 0., M. B. Robhins, K. Klenk, B. J. Blitvich, N. L. Marlenee, K. L.
Burkhalter, D. J. Gubler, G. Gonzalvez, C. J. Pena, A. T. Peterson, and N.
Komar. 2003. West Nile virus transmission in resident hirds, Dominican
Republic. Emerg. lnfect. Dis. 9:1299—1302.
Lai, C. 3., and T. P. Monath. 2003. Chinieric fiaviviruses: novel vacdines
against dengue fever, hek-home encephalitis, and Japanese encephalitis.
Adv. Virus Res. 61:469—509.
Lanciotti, R. S., J. T. Roebrig, V. Deubel, 3. Snsith, M. Parker, K. Stede, B.
Crise, K. E. Volpe, M. B. Crabtree, 3. H. Scherret, R. A. Hall, 3. S. Mac
kenzie, C. B. Cropp, B. Panigrahy, E. Ostlund, B. Schmitt, M. Malkinson, C.
Banet, 3. Weissman, N. Komar, H. M. Savage, W. Stone, T. McNamara, and
D. J. Gubler. 1999. Origin of the West Nile virus responsible for an outbreak
of encephalitis in the northeastern United States. Science 286:2333—2337.
Levenbook, 1. S., L J. Pelleu, and B. L. Elisberg. 1987. The monkey safety
test for neurovirulcnce of yellow fcvcr vaccines: the utility of quantitative
clinical evaluation and histological examination. 3. Biol. Stand. 15:305—313.
Marfin, A. A., and D. 3. Gubler. 2001. West Nile eneephalitis: art emerging
discase in the United States. Clin. Infect. Dis. 33:1713—1719.
Monath, T. P., 3. Arroyo, 1. Levenbook, Z.-X. Zhang, J. Catalan, K. Draper,
00
3
b
-i
(0
1F’
0
1
0,
CD
-
0
(1)
H
0
ED
0
1
0
G)
1F’
0
m
0
0
m
VOL. 78, 2004
23.
24.
25.
26.
27.
ChimeriVax-WEST NILE VACCINE PRECLINICAL EVALUATION
and F. Guirakhou. 2002. Single mutation in the flavivirus envelope prolein
hinge region incteases ncurovirulcnce for mice and monkeys hut decrcases
visccrotropism for monkcys: rclcvance to dcvclopment and safety tcsting of
live, attenuated vaccines. J. Virol. 76:1932—1943.
Monath, T. P., F. Guirakhoo, R. Nichols, S. Yoksan, R. Schrader, C. Mur
phy, 1’. Blum, S. Woociward, K. McCarthy, D. Mathis, C. Johnson, and P.
Bedford. 2003. Chimeric live, attcnuatcd vaccine against Japanesc cnccph
alitis (OsimeriVax-JE): phasc 2 clinical trials for safety and immunogenicity,
effect of vaccine dose and schedule, and memory response to challcnge with
inactivated Japanesc encephalitis antigen. 3. Infect. Dis. 188:1213—1230.
Monath, T. P., 1. Levenbook, K. Soike, Z..X. Zhang, M. Ratterree, K.
Draper, A. D. Barrett, R. Nichols, R. Weitzin, 3. Arroyo, and F. Guirakhoo.
2000. Chimeric yellow fever virus 17D-Japanese encephalitis virus vaccine:
dosc-response effectiveness and cxicnded safety testing in rhcsus monkeys.
3. Virol. 74:1742—1751.
Monath, T. 1’., K. McCarthy, P. Bedford, C. T. Johnson, R. Nichols, S.
Yoksan, R. Marchesani, M. Knauber, K. H. Welis, J. Arroyo, and F. Gui
rakhoo. 2002. Clinical proof of principle for ChimcriVax: recombinant live,
attenuated vaccines against flavivirus infections. Vaccine 20:1004—1018.
Monath, T. P., K. Soike, 1. Ixvenbook, Z. X. Zhang, 3. Arroyo, S. Delagrave,
G. Myers, A. D. Barrelt, R. E. Shope, M. Ratterree, T. J. Chambers, and
F. Guirakhoo. 1999. Recombinant, chirnacric live, attenuated vaccine
(ChimcriVax) incorporating the envelope gcncs of Japanesc encephalitis
(SAI4—14—2) virus and the capsid and nonstructural genes of yellow lever
(17D) virus is safe, immunogcnic and protective in non-human primates.
Vaccinc 17:1869—1882.
Nasci, R. S., N. Komar, A. A. Marfin, G. V. Ludwig, L D. Kramer, T. J.
Daniels, R. C. Falco, S. R. Camphell, K. Brookes, K. L. Gottfried, K. L
Burkhalter, S. E. Aspen, A. .1. Kerst, R. S. Lanciotti, and C. G. Moore. 2002.
Detcction of Wesi Nile virus-infcctcd mosquitoes and seropositive juvenile
hirds in the virinity of virus-positive dead birds. Am. J. Trop. Mcd. Hyg.
67:492—496.
12507
27a.National Research Council. 1996. Guide for the care and use of lahoratorv
animals. National Academy Prcss, Washington, DC.
28. Ni, H., G. J. Chang, H. Xie, D. W. Trent, and A. D. Barreit. 1995. Molccular
basis of attenuation of neurovirulence of Wild-type Japanese enccphalitis
virus strain SA14. J. Gen. Virol. 76:409—413.
29. Ni, H., K. D. Ryman, H. Wang, M. F. Saeed, R. Huil, D. Wood, P. D. Minor,
S. J. Watowich, and A. D. T. Barrett. 2000. Interaction of ycllow fcvcr virus
French neurotropic vaccine strain with monkey bruin: characterization of
monkey bruin mcmbrane receptor cscapc variants. J. Virol. 74:2903—2906.
30. Pletnev, A. G., M. S. Claire, R. Elkins, J. Speicher, B. R. Murphy, and R. M.
Chanock. 2003. Molecularly cngineered live-attenuated chimeric West Nilel
dengue virus vaccincs protect rhesus monkeys from West Nile virus. Virology
314:19(1—195.
31. Poidinger, M., R. A. Hall, and 3. S. Mackenzie. 1996. Molecular character
ization of the iapancse encephalilis serocomplex of the flavivirus genus.
Virology 218:417—421.
32. Rey, F. A., F. X. Heinz, C. Mandl, C. Konz, and S. C. Harrison. 1995. The
envelope glycoprotein from tick-borne encephalitis virus at 2 A reso!ution.
Nature 375:291—298.
33. Ryman, K. D., II. Xie, T. N. 1edger, G. A. Campbell, and A. D. Barrett, 1997.
Antigenic variants of ycllow lever virus with an altered ncurovirulence phe
notype in mice. Virology 230:376—380.
34. Sumiyoshi, H., G. H. Tignor, and R. E. Shope. 1995. Characterization of a
highly attenuatcd Japanesc cnccphalitis virus gencratcd from molecularly
cloned cDNA. J. Infect. Dis. 171:1144—1151.
35. Tesh, R. B., 3. Arroyo, A. P. Travassos Da Rosa, H. Guzman, S. Y. Xiuo, and
T. P. Monath. 2002. Efficacy of killed virus vaccine, live attenuated chimeric
virus vaccine, and passive immunization for prevention of West Nile virus
enccphalitis in hamster model. Emerg. lnfect. Dis. 8:1392—1397.
36. World Health Organization. 1998. Requiremcnts for yellow fever vaccine.
WHO. Expert Committee os Biological Standardization, forty-sixth report.
WHO Tech. Rep. Ser. 872:30-68.
0
0
0
0)
0
CD
0
-5,
-S
0
3
00
3
0
1
(0
0
0
-S
01
r\)
0
0
(‘3
H
0
112
0
-J
0
0
1
m
0
0
m
‘Dj(
‘
Veterinary Pathology Onhine
http://vet.sagepub.com/
Pathologic and Immunohistochemical Findings in Naturally Occurring West Nile Virus Infection in Horses
C. Cantile, F. Del Piero, G. Di Guardo and M. Arispici
Vet Pathol 2001 38: 414
DOl: 10.1354/vp.38-4-414
The online version of this article can be found at:
http://vet.sagepub.com/contentl38l4l4 14
Published by:
SAGE
http://www.sagepublications.com
On behalf of:
American College of Veterinary Pathologists, European College of Veterinary Pathoiogists, & the Japanese College of Veterinary
Pathologists.
Add,tional services and information for Veterinary Pathology Online can be found at:
Email Alerts: http:Ilvet.sagepub.com/cgilal erts
Subscriptions: http://vet.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
» Version of Record
-
Jul 1, 2001
What is This?
DownIoded fron, vet ongepob core ot MERCK & CO INC
0fl
Apni 5.2012
Vet Pathol 38:4)4—421 (2001)
Pathologic and Immunohistochemical Findings in Naturally
Occurring West Nile Virus Infection in Horses
C. CwriL, E DEL PIER0, G. Di GuAIuo,
AND
M. Aiusici
Dipartimento di Patologia Animale, Facolti di Medicina Veterinaria, Universitt degli Studi di Pisa, Pisa, Italy (CC, MA),
Departrnent of Pathobiotogy, School of Veterinary Medicine, New Bolton Center,
University of Pennsylvania, Kennel Square, PA (FDP); and
Istituto Zooprofilattico Sperimentale dclie Regioni Lazio e Toscana, Rome, Italy (GDG)
Abstract. The pathologic and peroxidase immunohistochemical features of West Nile flavivirus (WNV)
infection were compared in four horses from the northeastem United States and six horses from centra) Italy.
In all 10 animals, there were mild to severe polioencephalomyelitis ith small T Iymphocyte and lesser mac
rophage perivascular infiltrate, multifocal glial nodules, neutrophils, and occasional neuronophagia. Perivascular
hemorrhages, also noted macroscopicaily in two animals, were observed in 50% of the horses. In the four
American horses, lesions extended from the basai nuclei through the bram stem and to the sacral spinal cord
and were more severe than the lesions observed in the six Itaiian horses, which had moderate to severe lesions
mainiy in the thoracolumbar spinal cord and mild rhombencephalic lesions. WNV antigen was scant and was
identifled within the cytopiasm of a few neurons, fibers, glial celis, and macrophages. WNV infection in horses
is characterized by lesions with little associated antigen when compared with WNV infection in birds and some
fatal human infections and with other important viral encephalitides of horses, such as alphavirus infections
and rabies.
Key words:
Horses, immunohistochemistry, nervous system disease, polioencephalomyelitis, West Nile fia
vivirus.
West Nile virus (WNV) is a member of the genus
Flavivirus (family Flaviviridae) and was originally iso
lated from the blood of a woman with fever in the
West Nile province of Uganda in 1937.28 The virus is
primarily transmitted and amplified between avian res
ervoir hosts by many species of mosquitoes in enzo
otic transrnission cycles. During periods of favorable
ecologic and climatic conditions, virus-carrying mos
quitoes can transmit WNV to susceptible mammals
and birds, causing epidemics and/or epizootics of en
cephalitis in humans and horses and mortality in cer
tam domestic and wild birds. A recent review of the
3 In En
epidemiology of WNV has been published.’
rope, Africa, and Eurasia, WNV bas caused sporadic
cases and outbreaks in humans and horses since the
early 1960s,’° but the virus was isolated for the first
time in the United States in 1 999•3
West Nile virus was first recognized in Italy in a
1998 outbreak involving 14 horses from nine different
2 This epizootic occurred
farms in central Tuscany.
from August to October 1998, and 6 of the 14 affected
horses presented with sudden tetraparesis or parapa
resis, progressing to tetraplegia and recumbency; 8 an
irnals recovered without significant consequences.
Four of the six recumbent horses were euthanatized,
and the other 2 died spontaneously. No cases of dis-
ease in human beings and other domestic or wild an
imals were observed during the equine epizootic in
Tuscany. Fifty-seven cases of equine WNV encepha
loinyelitis were recognized from May to November
2000 in seven states in the northeastern United States.
A total of 23 horses died or were euthanatized because
4 Prelirninary reports sug
of severe neurologic signs.
gested that a cluster of 25 cases of equine neurologic
disease coinciding with the 1999 human cases of fatal
viral encephalitis in New York City were most likely
34 Although it is still unclear
due to WNV infection.
how many horses developed WNV encephalomyelitis
in 1999 in the counties nearby New York City, the
virus spread rapidiy in 1 year to the equine populations
of the adjacent coastal states, and rapid additional virus
diffusion is expected.
The clinical and neuropathologic features of spon
taneous ‘%VNV infection in the horse have been re
2 However, the lesions in horscently reported in detail.
es from the northeastern United States have not been
reported and the tissue distribution of WNV in horses
bas not previously been described. In this comparative
study, which includes horses from Italy and the United
States, we evaluated the lesions and the immunohis
tochemical distribution of WNV antigen in the ner
vous systern and major organs of affected horses using
4)4
Dowr,Ioode from v,t,ogopob oor’ 01 MERCK & CD INC
0fl
Apni 5.2012
Table 1.
415
West Nile Virus in Horses
Vet Pathol 38:4, 2001
Breed, age, gender, and origin of the 10 horses with WNV encephalomyelitis.
Horse No.
Breed
1
2
3
4
5
6
7
8
9
10
F = female; M
=
Age
Standardbred
2 years
Standardbred
5 years
Standardbred
6 years
Standardbred
16 months
Wielkopolski
12 years
Standardbred
9 years
Thoroughbred
30 years
Quarter Horse
18 years
Thoroughbred
8 years
Thoroughbred
4 years
male; MC = male castrated.
serologic, virologic, and molecular methods. We also
described the phenotypic characterization of the in
flammatory celi infiltrates.
Materials and Methods
Complete postmortem examinations were carried out on
six horses from Tuscany, central Italy. The horses were all
females and ranged in age from 16 months to 12 years (horse
Nos. 1—6; Table 1). These horses were examined during an
epizootic of WNV infection in farms close to a ivetland pop
ulated with numerous migratory birds and mosquitoes.
2 Post
mortern examinations were also completed on 4 horses from
the East Coast of United States (New Jersey, horse Nos. 7,
8, 10; New York state, horse No. 9; Table 1), that were
submitted to the Department of Pathobiology, Section of
Large Animal Pathology, New Bolton Center, University of
Pennsylvania. The spinal cord of horse No. 8 was not ex
amined because this animal was submitted with a presump
tive diagnosis of viral encephalitis and had not been vacci
nated for rabies.
West Nile virus infection in horses from Italy was previously
confirmed by virus isolation, reverse transcription polymer
ase chain reaction (RT-PCR) and serologic examination
(horse Nos. 1_6).2 Virus isolation on RK-l3Ky cultured celis
and identification with specific monoclonal antibodies rec
ognizing a WNV structural glycoprotein (glycoprotein E)
were conducted with a Pool of fresh tissues (cerebral cortex,
bram stem, cerebellum, spinal cord) from the four horses
from the northeastern United States. Further virus identifi
cation for these tissues was ohtained using a previously
1 slightly modified to achieve
described RT-PCR technique’
resuits directly from pools of fresh tissues (E. Ostlund, per
sonal communication). Representative tissue samples of ma
jor organs and the w’hole centra! nervous system (CNS) ‘ere
fixed in 10% neutral buffered formalin solution and pro
cessed for histology. Tissue sections were stained with he
rnatoxylin and eosin (HE), periodic acid—Schiff, Luxol fast
blue, and crystal violet for Nissl substance and were pro
cessed with indirect immunoperoxidase staining using an au
tomated slide processing unit (Autostainer Immunostaining
System S3400, DAKO Corp., Carpinteria, CA). Tissue sec
tions were mounted on Probe-On (Sigma Diagnostics, St.
Louis, MO) glass slides coated with poly-L-lysine and de-
Gender*
F
F
F
F
F
F
MC
MC
F
M
Location
Outcome
Tuscany, Italy
Tuscany, Italy
Tuscany. Italy
Tuscany, Italy
Tuscany, Italy
Tuscany, Italy
New Jersey
New Jersey
New York
New Jersey
Euthanasia
Euthanasia
Deceased
Euthanasia
Deceased
Euthanasia
Euthanasia
Euthanasia
Euthanasia
Euthanasia
paraffinized using 1-lemo-De (Fisher Scientific, Ni). Endog
enous peroxidases ‘ere inactivated with H
. Tissues were
0
2
incubated in Proteinase K (DAKO) for 5 minutes at room
temperature. The primary antibody was a murine polyclonal
antibody recognizing specific epitopes of WNV. The anti
body was diluted 1:50 in 4% diluent (DAKO) and incubated
for 30 minutes at room temperature. The secondary antibody
was a universal goat anti-mouse immunoglobulin conjugated
to a peroxidase-labeled polymer (EnVision+TM, DAKO),
which was incubated for 20 minutes at room temperature.
The chromogen 3,3-dianilnobenzidine-4HCI, was incubated
for 3 minutes at room temperature. The sections were coun
terstained using Mayer’s hematoxylin (DA.KO), dehydrated,
and coverslipped. The WNV positive control tissues were
wild goose cerebrum, cerebellum, and heart from which
WNV had been isolated and identitied using the techniques
described above. WNV-free avian and maminalian tissues
were used as negative controls, as were equine nervous tis
sues containing rabies virus, encephalitic togaviruses, and
equine herpesvirus 1 (EHV-1). Paired tissues samples were
also stained using various separate primary mouse monoclo
aal and rabbit polyclonal antibodies, including specific re
agents for equine herpesviruses, equine arteritis virus, and
eastern equine encephalitis alphavirus, that did not react with
WNV. The absence of nonspecific binding of the secondary
antibody was also evaluated by omitting the primary anti
body.
The leukocyte populations infiltrating the CNS were ex
amined using antibodies recognizing CD3 of T ceils (rabbit
polyclonal antibody, dilution 1:100, DAKO), BLA.36 of B
cells (murine monoclonal antibody, dilution 1:60, DAKO),
and MAC 387 of macrophages (murine monoclonal anti
body, dilution 1:1600, DAKO). Pretreatment included pro
teinase K (DAKO) for 5 minutes. The identity of celis la
beled by each antibody was confirmed by the presence of
morphologic features characteristic of each celi type. Posi
tive controls were tissues from normal horses containing the
appropriate cell types.
Resuits
Histologic lesions and peroxidase immunohisto
chemistry results are reported in Table 2. Results of
clinical examination and neuropathology of the horses
Downloaded 1,0,,, vetoagepob oor,, at MERCK & CD INC en April 5.2012
Cantile, Dcl Piero, Di Guardo, and Arispici
416
Table 2. Distribution and degree of histologic Iesions* and WNV antigen
encepholmyelitis.
Vet Pathol 384, 200t
in the CNS
of
the
10 horses with WNV
Spinal Cord
Pons
+
+
++
++
++
—
—
++
++
++
+
+
Cerebral
Cortex
Thalamus
Lesions
Virus
—
+1—
+1—
+
—
Lesions
Virus
+
Horse No.
Cervical
Thoracic
Lunibosa
cral
++
++
++
++
+++
++
+++
++
+++
+
NA
NA
NA
NA
NA
NA
+
++
+
++
++
+
++
++
+
+
+
+
++
+
++
+
+
+
++
+
++
+
Cerebellar Medulla
Oblongata
Cortex
Mesencephalon
Basal
Nuclei
2
-
+
-
-
-
-
3
Lesions
Virus
—
—
—
+
+1—
—
—
—
+
+
+
+
—
—
4
L,esions
Virus
—
—
+1—
—
+I
—
—
—
—
—
5
Lesions
Virus
—
—
+1—
+1—
+
+
+
+
++
+
+
+
+
+
+
—
—
—
—
—
—
6
++
++
Lesions
Virus
+
Lesions
Virus
—
Lesions
Virus
+/—
-
++
+
+++
+
++
++
++
++
Lesions
Virus
—
—
++
+
+
+
+
+
++
Lesions
—
+1’-
+/—
+
+
-
-
—
—
—
—
+/—*
—
++
+
++
+
++
+
++
+
—
—
++
+
++
+
+++
+
+
+++
++
NA
NA
NA
NA
NA
NA
+
++
+
+
NA
NA
NA
NA
+
+
+
NA
NA
7
—
+±
+
++
+
8
+I—
—
9
—
—
—
10
Virus
*
Degree of infiammatory changes (lesions); —
=
none; +1—
—
occasional; +
-
-
-
-
=
=
mild; ++
=
nioderate; +++
available.
= negative; + = single ccli positivity;
not available.
Presence of perivascular hemorrhage.
§ Occasional presence of infiammatory les ions in some cortical areas associated with leptomeningitis.
t lnimunohistochemical results, amount of WNV antigen (virus): —
NA
+
+
NA
NA
severe; NA not
muitiple celi positivity;
=
involved in the 1998 WNV epizootic in central Italy
2 Infection in all horses
have been reported elsewhere.
was characterized by mild to moderate, multifocal
lymphocytic polioencephalomyelitis with constant in
volvement of the ventral and lateral horns of the tho
racic and lumbar spinal cord and associated with mod
erate to severe hemorrhage in some horses. Slight to
moderate Iymphocytic and monocytic infiammation
and scattered foci of microgliosis were observed in the
medulla oblongata and pons and, to a lesser extent, in
the basal nuclei, thalamus. and mesencephalon.
In one horse from the northeastern United States
(horse No. 7), lesions were macroscopically evident,
with petechiae sparsely distributed throughout the en
tire rhon3bencephalon and extending multifocally
through the entire spinal cord (Fig. 1). These petechiae
were especially prominent within the thalamus, the
caudal bram stem, and the ventral horns of the spinal
cord. All four Arnerican horses showed microscopic
changes similar to those of the Italian horses. These
changes were characterized by multifocal mild to se
vere perivascular cuffs comprised of lymphocytes and
OownI000ed Iroro vet.ragepb con, o MERCK & CO INC en ApnI 5.2012
West Nile Virus in Horses
Vet Pathol 38:4, 2001
417
‘t
5-
t.
?‘;--.
..,.
•‘“
:-
:t
‘‘
..r
1
)
0’t4
S
,
•.
‘
‘:
t,
-.
,
t,
‘“5
,
t
;.,
t
sta’
.
8’
‘2
•
.,,.
;t
t
-
;
-
‘t!
t
-
.4
‘t’,.
‘4
‘.
“.r
t’”’
q
0
t
,._a,
-.t•
‘
tt
..
.5’
)‘
;
S,.
AØ
4.
4’
‘•t
*
“t’
—4
1
S
.1
3
S
1
.,
a
t
,
h
t
‘‘t
Gt
•
—
t
“4
•
.—-—t.
0
•“
•
_t
4
4
•4
•
t..
-
t.
4
S.
S
•
t
S
*
4
4
.r,
S
1
‘,fr’
t
t,
;,
•:
t.
-
S
9*
•
S
t
1•
4
t
4.
*
t
•
•.
.:
,b
‘.
.
t—.,.
t
—
4
—‘t
b
;.,....
:«
t.’
•
‘
—
4
4,
4*
t4%•
t
4
4%,
t
t—
#
4
4
4,
t
*4
Fig. 1. Left thalamus (left) and cervical spinal cord (right); horse No. 7. Severe, multifocal, acute hen1orrhages mainly
involve the gray matter. The ventral and medial edges of the thalamus have been trimmed, and the tissue is partially fixed
in buffered formalin. Bar = 2 cm.
Fig. 2. Thoracic spinal cord; horse No. 1. Lymphocytic cuffs and infiammation of the neuropil with neuronal degen
eration and neuronophagia (arrow). HE. Bar 200 p.m.
Fig. 3. Caudal bram stem; horse No. 8. Severe perivascular neutrophilic infiltrate with glial celis and a degenerate,
shrunken, neuron surrounded by glial celis (bottom left). HE. Bar = 65 Ii.m.
Fig. 4. Caudal bram stem; horse No. 7. Margination of intravascular leukocytes, diapedesis, hyaline degeneration of
the arteriolar wall, and prominent perivascular hemorrhage. IIE. Bar = 65 pm.
Fig. S.
Medulla oblongata, reticular substance; horse No. 8. A prorninent swollen axon and several macrophages (bottom
left) contain WNV antigen. Peroxidase immunohistochemistry and hematoxylin. Bar = 65 Ji.m.
Fig. 6. Basal nuclei; horse No. 1. Neuron infected with WNV surrounded by glial ceils, lymphocytes, and a few
macrophages. Neuronal cytoplasm, axon and a few glial celis contain WNV antigen. Peroxidase immunohistochemistry and
hematoxylin. Bar
65 p.m.
Downloaded lrorrr vet .agepub cor,, at MERCK & CO INC orr April 5,2012
418
Cantile. Del Piero, Di Guardo, and Arispici
a few macrophages, with mild infiltration of lympho
cytes and gliosis in the adjacent neuropil (Fig. 2). In
the most severely affected areas, neuronal degenera
tion was prominent and characterized by central chro
matolysis, cytoplasmic swelling, or ceil shrinkage.
Small, scattered microglial foci, occasional neurono
phagia, and presence of small necrotic areas cornprised
of macrophages, neutrophils and cellular debris were
also observed (Fig. 3). In all horses from the north
eastern United States, these lesions involved the basal
nuclei, gray matter of thalamus, midbrain, lower bram
stem, and ventral and lateral homs of the spinal cord.
Scattered and small microglial foci were sometimes
observed in the white matter of the spinal cord. Swol
len axons (spheroids) were also noted in the most se
verely affected spinal cord segments. Six of the 10
horses had a prom inent diapedesis of leukocytes, with
mild to severe perivascular hemorrhage, degenerative
changes and occasional necrosis of the vessel wall
with presence of a few neutrophils; within the CNS
blood vessel lumina, there were marginating small
lymphocytes and macrophages (Fig. 4).
Mild focal infiammation of cerebral and cerebellar
cortical areas associated with mild leptorneningitis was
occasionally observed. Choroid plexi, ependyma, and
the pituitary and pineal glands were not affected. No
lesions were detected in the remaining areas of the
CNS, eyes, peripheral ganglia, or peripheral nerve
samples. Age-related lesions in sorne older horses in
cluded neuronal lipofuscinosis, vascular siderocalci
nosis, choroid plexus hyalinosis, and cholesterinosis.
A small adenoma of the pars intermedia of the pitui
tary gland was observed in horse No. 9. Lesions in
extraneural tissues of horses from the United States
included mild scattered hemorrhages in the kidney me
dulla and atrophy of spienic lyniphoid follicles (horse
No. 7) and mild, focal and nonsuppurative myocarditis
(horse No. 9).
In sites of iiiflainmatory lesions, WNV antigen was
mainly localized within the gray matter and had a fine
ly granular appearance within the cytoplasm of a few
morphologically normal and degenerate neurons.
WNV antigen was also present in a large number of
morphologically normal nerve fibers, axonal hillocks,
glial cells, and spheroids of the medulla oblongata and
spinal cord in all horses (Fig. 5). In the glial foci and
areas of neutrophilic infiltration, small amounts of
WNV antigen were contained within the cytoplasm of
glial cells, macrophages, and occasional neutrophils.
Occasionally, morphologically normal neurons dif
fusely harbored intracytoplasmic WNV antigen (Fig.
6) and viral antigen-positive glial cells also surrounded
some neuronophagic neurons. No viral antigen was de
tected within the peripheral nervous system and extra
neural tissues.
Vet Pathol 384, 200t
Perivascular and leptom eningeal infiammatory infil
trate was composed almost exclusively of small T lym
phocytes labeled for CD3 antigen. Occasionally, the
perivascular infiltrates were composed purely of mac
rophages labeled for MAC 387, and rare macrophages
contained intracytoplasmic WNV antigen. No BLA.36
positive leukocytes were identified.
Discussion
The histologic lesions and clinical sigils observed in
the affected horses were siinilar to those in the only
previously described case of natural equine WNV in
fection where a detailed pathologic examination was
carried out.
9 Lesion morphology and distribution were
typical of polioencephalomyelitis, with prevalent in
volvement of lower bram stem and ventral horns of
the thoracolumbar spinal cord, and were distinct from
those of other equine encephalomyelitides.
2 The results
obtained in the present study indicate that WNV ex
hibits a pronounced, 1f not exclusive, CNS tropism in
the horse. In all cases, the quantity of WNV antigen
was scant when compared with the extent of the in
flammatory lesions. Perivascular infiltrate and hemor
rhage were frequent findings, but no evidence of direct
viral infection of the vessel wall was detected by im
munohistochemistry.
The lack of apparent WNV antigen in extraneural
tissues of the horses in this study is distinctly different
from the widespread production of WNV antigen seen
in infected birds
° and indicates that WNV infection in
3
horses is largely limited to CNS infection. Although
some virus infection of other tissues may have oc
curred early in the course of disease and been cleared
prior to death, the lack of antigen and virus-associated
lesions in extraneural tissues makes this possibility un
likely. The Italian horses were examined 2 and 9 days
following the onset of clinical signs,
2 and most Amer
ican horses were euthanatized shortly after the onset
of clinical disease; thus, limited numbers of virions
may be sufficient to stimulate a significant CNS in
flammatory response. The disassociation of virion den
sity from lesion severity suggests an immunopatholog
ic mechanism in the pathogenesis of WNV disease in
horses.
Different genetic variants of WNV may exhibit dif
ferent pathogenicity for mammals and avian species.
The strain of WNV isolated in the United States and
in lsrael is highly pathogenic in many species of
, 30 whereas other WNV strains have not been
3
birds,t
associated with avian disease in nature. Strain varia
tion also has been associated with differences in path
ogenicity of WNV for ceil cultures and mice,’
8 and
differences in virulence for human beings has been
L Neuro
2
denionstrated for a number of viral strains.
virulence and neuroinvasiveness of encephalitogenic
Dowrloaded froo vetagepb core al MERCr< & CO INC on Apni 5,2012
Vet Pathol 38:4, 2001
West Nile Virus in Ilorses
flaviviruses has been associated with changes of the
amino acid sequence of viral glycoprotein E and with
mutations in the nonstruetural or untranslated regions
of the fiavivirus genome.’
5 Our finding that horses
from the United States exhibited more widespread le
sions in the bram than did horses from Italy suggests
that the US strain of WNV is highly pathogenic in
horses as well as birds. The Italian isolate is consid
ered closely related to the WNV isolate from a Moj
occan equine epizootic in 1996.’ No cases of avian or
human disease have been associated with that partic
ular WNV isolate.
32
West Nile virus experimental infection has been at
tempted in horses, mules, sheep,
° donkeys,
3
24 dogs,’
and pigs,” resulting in low-level and transient viremia
that was no longer detectable in horses 4—8 days fol
lowing inoculation.’
9 Mild pyrexia was observed 3—6
days following WNV inoculation. A second mild py
retic phase was identified only in those animals that
developed neurologic signs 5—22 days after inocula
0 Virus reisolation from different areas of the
tiofl.19,2
CNS, in relation to the route of inoculation and du
ration of the infection, suggested a descending neural
kinetic of viral infection.’
9 The onset, progression, and
outcome of the disease depends on a number of fac
tors, including species and route of inoculation,
33 in
fective virus burden,’
2 age,
823 and, as observed in hu
23 In the horses of this report, no
mans, health status.
influence of age was readily discernible, although oth
er host factors may influence the course and outcome
of flavivirus infection, ineluding immune responses
and genetic factors.
26 In the horses of this study, no
other important diseases were detected, suggesting that
a concurrent disease does not play a major role in the
pathogenesis of WNV infection in horses.
The route of neuroinvasion by flaviviruses remains
controversial. It has been suggested that low-level vi
remia in clinical hosts is inadequate to sustain endothe
hal ceil viral replication, resulting in virus budding on
the neuroparenchymal side. More likely, encephalito
genie arboviruses could invade the olfactory neuroepi
thelium during the viremic phase, subsequently infect
ing the olfactory neurons by retrograde axonal trans
6 The negative finding of WNV antigen in the en
port.’
dothelial celis from all horses investigated here is
consistent with this hypothesis. Once in the CNS, WNV
may spread rapidiy throughout the neuropil and may
rephicate in particularly sensitive areas, such as the hip
pocampus in mice, and the thalamus, substantia nigra,
and cerebellum in monkeys and human beings.’
22
’
7
Contrary to findings in mammals, both spontaneous
WNV infection of several bird species
° and experi
3
mental infection in chickens
25 have been associated
with lesions in multipie organs and, in crows and other
wild birds, presence of large amount of viral antigen
419
within either CNS and extraneural organs. Restricted
locahization of viral antigen within neurons, neuronal
processes and necrotic areas, mainly of the bram stem
and spinal cord, has been demonstrated by immuno
histochemistry in an investigation of human WNV en
27
cephalitis.
Viral distribution varies considerably arnong differ
ent equine encephalomyelitides. in eastern equine en
cephahitis, alphavirus antigen is abundant and locahized
within the cytoplasrn of neurons, glial cells, and ex
traneural celis, such as smooth muscle, myocardial fi
bers, dendritic cells.
7 In horses with rabies, rhabdovi
rus antigen tends to be abundant and widespread with
in the gray matter, including peripheral ganglia and
retina, being located within the cytoplasm of neuronal
ceil bodies, nerve fibers, and some glial ceils.
6 In ra
bies and alphavirus encephalitides, there is also prom
inent cerebral cortical involvement.
’ EHV-l induces
6
ischemic myeloencephalopathy with perivascular lym
phocytic cuffing secondary to endothelial necrosis and
thrombosis associated with vasculitis, and small quan
tities of EHV- 1 antigen are expressed within nucieus
and cytoplasm of small muscular artery endothehium,
myocytes, and pericytes, mainly in the spinal cord.
5
Considering the sporadic occurrence of WNV-as
sociated disease in horses, moderate lesions, short vi
remia, and limited amount of viral antigen detected in
tissues, horses may be incidental, dead-end hosts of
this flavivirus. Consequently, horses most likely do not
play a significant role in the epidemiology of WNV
and may not pose a major risk to human health. How
evei special protection should be used during the dis
section and examination of the CNS of WNV-infected
horses. The himited tissue distribution of WNV antigen
apparent in the horses of this study should be taken
into account during postmortem examination. All six
horses from Italy had localized lesions and viral antigen distribution almost exclusively himited to the low
er bram stem and thoracolumbar spinal cord, with mild
occasional involvement of the cerebrum. Although re
moval of the spinal cord from horses infected with
neurotropic viruses can be potentially dangerous to the
prosector and current regulations discourage the re
moval of the spinal cord from possible rabies cases,
our findings demonstrate that failure to examine the
spinal cord limits the ability to accurately identify
WNV infection in horses.
In WNV-infected horses, how-level viremia, anti
body production, and sub chinical illness seern to be
the most common manifestations. Serologic investi
gations in a number of WNV equine disease outbreaks
revealed that many horses had WNV antibodies, al
though only a few developed chinical signs. In clini
cally affected horses, a very serious and hife-threat
Downloeded trom vetsgepub.cormm al MERCK & CO INC orm ApmO 5,2012
420
Cantile, Del Piero, Di Guardo, and Arispici
Vet Pathol 38 4, 2001
C. Clinical, pathological, immunohistochemical, virolog
ical tindings of eastern equine encephalitis in two horses.
Vet Pathol 38:451—456, 2001
8 El Dadah AH, Nathanson N: Pathogenesis of West Nile
virus encephalitis in mice and rats. 11. Virus multiplica
tion, evolution of irnrnunofluorescence, and development
of histological lesions in the bram. Am J Epidemiol 86:
776—790, 1967
9 Guillon JC, Oudar J, Joubert L, Hannoun CL: Lésions
histologiques du système nerveux dans l’infection a virus
West Nile chez le cheval. Ann Inst Pasteur 114:539—550,
1968
10 Hubâlek Z, Halouzka i: West Nile fever—a reemerging
mosquito-borne viral disease in Europe. Ernerg Infect
Dis 5:643—650, 1999
11 Ilkal MA, Prasanna \Ç Jacob PG, Geevarghese G, Ba
nerjee K: Experimental studies on the susceptibility of
domestic pigs to West Nile virus followed by Japanese
Acknowledgements
encephalitis virus infection and vice versa Acta Virol
We are very grateful to the scientists ‘ho helped us. Dr.
38:157—L61, 1994
Vincent Deubel (Centre de Recherche Mérieux Pasteui
12 Joubert L, Oudar J, Hannoun C, Chippaux M: Repro
Lyon, France) provided one of the anti-WNV antibodies uti
duction expérimentale de la méningo-encéphalomyélite
lized in this study and identified the WNV of the Tuscany
du cheval par l’arbovirus West Nile. III. Relations entre
epizootic, Dr. Eileen N. Ostiund (National Veterinary Ser
Ja virologie, Ja sérologie et l’évolution anatomo-clinique.
vices Laboratories, Ames, JA) identified the Arnerican WNV
Conséquences épidérniologique et prophylactiques. Bull
isolates, Dr. D. E. Swayne (Southeast Poultry Research Lab
Acad Vét 44:159—167, 1971
oratory, US Department of Agriculture. Athens, GA) pro 13 Koinar N: West Nile virus encephalitis. Rev Sci Tech
vided the WNV avian positive controls, and Dr. Pamela Wil
0ff Int Epizoot 19:166—176, 2000
kins (Department of Clinical Studies, New Bolton Center) 14 Lanciotti RS, Roebrig JT, Deubel V, Smith J, Parker M,
provided assistance in manuscript preparation. We thank Dr.
Stede K. Crise B, Volpe KE, Crabtree MB. Scherret JH,
Perry Habecker, chief of the pathology service at New BolHall RA. MacKenzie JS, Cropp CB. Panigrahy B, Ost
ton Center, Dt Daniela Ennulat (Schering-Plough), and
lund E, Schmitt B, Malkinson M. Banet C, Weissman J,
Ralph Conti (New Bolton Center) for necropsy help. We are
Komar N, Savage HM, Stone W McNamara T, Gubler
particularly grateful to Mrs. Jaqueline Ferracone (New BolDi: Origin of the West Nile virus responsible for an out
ton Center) and Mrs. Lisa Baroncini (Dipartirnento di Pa
break of encephalitis in the northeastern United States.
tologia Animale, University of Pisa) for superb assistance
Science 286:2333—2337, 1999
with irnrnunohistocheniistry and histology.
15 McMinn PC: The molecular basis of virulence of the
encephalitogenic flaviviruses. J Gen Virol 78:27 1 1—
References
2722, 1997
1 Blackburn NK, Reyers l Berry WL, Shepherd AJ: Sus 16 Monath T1 Cropp CB, Harrison AK: Mode of entry of
a neurotropic arbovirus into the central nervous system.
ceptibility of dogs to Vest Nile virus: a survey and path
Reinvestigation of an old controversy. Lab Invest 48:
ogenicity trial. J Comp Pathol 100:59—66, 1989
399—410, 1983
2 Cantile C, Di Guardo G, Eleni C, Arispici M: Clinical
and neuropathological features of West Nile virus equine 17 Nathanson N, Gittelson AM, Thind IS: Histological stud
ies of the rnonkey neurovirulence of group B viruses.
encephalornyelitis in italy. Equine Vet J 32:31—35, 2000
Am 1 Epidemiol 85:503—517, 1967
3 Centers for Disease Control: Update: West Nile virus en
cephalitis—New York, 1999. Morbid Mortal Weekly 18 Odelola HA, Fabiyi A: Biological characteristics of Ni
gerian strains of West Nile virus in mice and cdl cul
Rep 48:944—946, 1999
tures. Acta Virol (Praha) 21:161—164, 1977
4 Centers for Disease Control: Update: West Nile Virus
activity—eastern United States, 2000. Morbid Mortal 19 Oudar 1, ioubert L, Hannoun C, Corniou B: Reproduc
tion expérimentale de Ja méningo-encéphalomyélite du
Weekly Rep 49:1044—1047, 2000
cheval par l’arbovirus West Nile. 1. Etude virologique et
5 Del Piero F Dubovi EJ. The diagnosis of equine herpes
sérologique. Bull Acad Vét 44:107—122, 1971
virus 1 (EHV-1) abortion and neonatal infection withern
phasis on immuno-peroxidase histochemical findings. 20 Oudar J, Joubert L, Lapras M, Guillon JC: Reproduction
expérimentale de la méningo-encéphalomyélite du chev
Vet Pathol 35:444, 1998
al par l’arbovirus West Nile. II. Etude anatomo-clinique.
6 Dcl Piero E Wilkins PA, de Lahunta A, Trimarchi C,
Bull Acad Vét 44: 147—158, 1971
Dubovi Ei. Rabies in horses in New York state: clinical,
pathological, immunohistochemical and virological find 21 Parks JJ, Ganaway JR, Price WH: Studies on immuno
logic overlap among certain arthropod-borne viruses. III.
ings. Proc int Conf Equine Infect Dis 8:29 1—296, 1998
A laboratory analysis of three strains of West Nile virus
7 Del Piero F Wilkins PA, Dubovi Ei, Biolatti B, Cantile
ening disease develops, with a fatality rate of 43—
45% 13
WNV should be considered an important pathogen
for horses, capable of inducing significant neurologic
disease in temperate regions, where enzootic infection
cycles may develop and be n3aintained.’° The paucity
of WNV antigen in equine nervous tissues indicates
that multiple sections of bram and spinal cord labeled
by immunohistochemistry should be examined, and
negative resuits should be interpreted with significant
caution. The use of histopathology, peroxidase im
munohistochemistry, virus isolation, and RT-PCR
techniques will increase the sensitivity for WNV iden
tification, aliowing further characterization of this in
creasingly important disease.
Downloeded
tror,
et agepab corn al MERCK & CO INC
0r
Aprl 5,2012
c
Vet Pathol 38 4 2001
West NjIe Virus in Horses
421
29 Snook CS, Hyman SS, Dcl Piero F Palmer JE, Ostlund
EN, Barr BS, Desrochers AM, Reilly LK. West Nile vi
rus encephalomyelitis in eight horses. J Am Vet Med
Assoc (in press).
30 Stede KE. Linn MJ, Schoepp Ri, Komar N, Geisbert
TW, Manduca RM, Calle PF Raphael DL, Clippinger
Smith 1, Lanciotti RS, Panella NA, Mc
TL, Larsen
Namara TS: Pathology of fatal West Nile virus infections
in native and exotic birds during the 1999 outbreak in
infection of Egyptian equines with West Nile virus. Ann
New York City, New York. Vet Pathol 37:208—224, 2000
Trop Med Parasitol 57:415—427, 1963
31 Taylor RM, Work TH, Hurlhut HS, Rizk FA: A study of
Senne DA, Pedersen JC, Hutto DL, Taylor WD, Schmitt
the ecology of West Nile virus in Egypt. Am J Trop Med
BJ, Panigrahy B: Pathogenicity of West Nile virus in
Hyg 5:579—520, 1956
chickens. Avian Dis 44:642—649, 2000
32 Tber AA: West Nile fever in horses in Morocco. Bull
Shellam GR, Sangster MY, Urosevic N: Genetic control
0ff Int Epizoot 108:867—869, 1996
of host resistance to flavivirus infection in animals. Rev
33 Weiner IJ Cole GA, Nathanson N: Experimental en
Sci Tech 0ff Int Epizoot 17:23 1—248, 1998
cephalitis following peripheral inoculation of West Nile
Shieh WJ, Guarner J, Layton M, Fine A, Miller J, Nash
virus in mice of different ages. J Hyg (Lond) 68:435—
D, Campbell GL, Roehrig JT. Gubler DJ, Zaki SR: The
446, 1970
role of pathology in an investigation of an outbreak of
Wilson AJ, Meade B, Varty TJ, Hutto D, Gidlewski T,
34
West Nile encephalitis in New York. 1999. Emerg Infect
Gregg D: Pathological and epidemiological investigation
Dis 6:370—372, 2000
of West Nile virus disease in equines in Long Island,
Smithburn KC, Hughes TI Burke AW, Paul JH: A neu
New York. Proc Annu meet Am Assoc Vet Lab Diagn,
rotropic virus isolated from the blood of a native of
43:5. 2000.
Uganda. Arn J Trop Med Hyg 20:47 1—492, 1940
which have been studied in human cancer patients. Am
J Hyg 68:106—119, 1958
22 Reyes MG, Gardner JJ, Poland JD, Monath TP: St. Louis
encephalitis. Quantitative histologic and irnmunofluores
cent studies. Arch Neurol 38:329—334, 1981
23 Sampson BA, Ambrosi C, Charlot A, Reiber K. Veress
JF Armbrustmacher V: The pathology of human West
Nile Virus infection. Hum Pathol 31:527—53 1, 2000
24 Schmidt JR, El Mansoury HK: Natura! and experirnental
25
26
27
28
Request reprints from Dr. F Del Piero, Department of Pathobiology, School of Veterinary Medicine, New Bolton Center,
University of Pennsylvania. 382 West Street Road, Kennett Square, PA 19348-1692 (USA). E-mail: fdpvet.upenn.edu.
Downloadod frorrr ver oogepeb corrr al MERCI(
CO INC on April 5. 2012
Study on the stability of 17D-204
yellow fever vaccine before and after
stabilization
D.K. Sood*, R.K. Aggarwal, S.B. Sharma, J. Sokhey and H. Singh
(
To monitor the parameters controlling yellowfever vaccine production, eighi diijrent lots
produced wit/zout stabilizers were studied. It was Jund that the freeze-thaw cycle did not
have any adverse effect on virus inJctivity and the mean loss in virus titre during
vophilization was 0.51 1og
10 ml’. A thermodegradation study after storage of vaccine
al different temperarures sho wed that the vaccine did not pass t/ze accelerated stahility
test. To stahilize the 1 7D-204 substrain vaccine, four stabilizers with dfferent sugars and
amino acids were examined. The optimum time for addition of the stabilizers was found
to be during homogenization of infected embryos. An accelerated stability test at 37°C
3 kept the vaccine stable for up to 4 weeks, whereas
, S, and S
1
indicated that stabi/izers S
3 was found
. Stabilizer S
4
the vaccine was only stable for up to 3 weeks in stabilizer S
sta tistically to be the best. Reproducibility in production meihodology was established by
.
3
preparing and resting more baiches of the vaccine using stabilizer S
Keywords: Yeilow fever; stability; temperature
Since 1938, when Smith and co-workers’ described the
technique of 1 7D yellow fever vaccine production and
methods employed in safety measures and potency assay,
large numbers of people residing in or visiting endemic
areas have been vaccinated. Although the vaccine is found
to be a safe immunogenic agent and its production
procedures have been well established, the literature is
stili sparse on the infectivity titre of the virus during
various stages of its production. Therefore, in the first
part of the study a total of eight different vaccine lots
prepared without stabilizers have been analysed in order
to establish the necessary parameters to improve vaccine
quality control.
The first part of the study and the findings of other
workers have shown that even in a Iyophilized form, the
vaccine is readily inactivated when exposed to ambient
. This poses problems as the vaccine is
2
temperature
mainly used in developing and tropical countries where
transportation and refrigeration facilities are scarce.
Despite precautions taken, it may lose potency before
arrival at the place where it is to be used. Therefore,
stability criteria for many freeze-dried viral vaccines such
as measles, Japanese encephalitis and rabies have been
laid down by WHO
. Similarly, it bas been suggested
3
that yeltow fever vaccine should retain the minimum titre
of 1000 mouse LD
50 per dose when exposed to 37°C for
10
14 days and the drop in titre should be 1.0 1og
(Ref. 4).
YeIlow Fever Vaccine Section, Central Research Institute,
Kasaul -173 204 (H.P.), India. *To whom correspondence
should be addressed. (Received 2 June 1992; revised 25 July
1992; accepted 10 December 1992)
0264—410XJ93/1 1/1124-05
© 1993 Butterworth-Heiflemaflfl Lid
1124
Vaccine, Vol. 11, lssue 11, 1993
The effect of chemicals on the heat resistance of this
vaccine has been studied by various workers
. This
5
new vaccine is found to be potent when studied for
accelerated stability behaviour. In view of these findings
efforts have been made to improve the stability behaviour
of the present 17D yellow fever vaccine using one or more
stabilizers with the aim of finding the appropriate
formulation and time of adding the stabilizer to result in
a vaccine meeting the aforesaid stability requirement
suggested by WHO.
MATERIALS AND METHODS
The methods used for the preparation of the 17D-204
substrain as a yellow fever vaccine were essentiafly those
meeting the international requirements’°. The seed lot
1 was adopted and all the
system certified by WHO’
prepared
at passage level 232 using
lots
were
vaccine
conventional embryonated eggs from white Leghorn
hens. Briefly, the infected chicken embryos were
harvested, homogenized to obtain 33% emulsion in
distilled water or in stabilized solution, distributed,
sheli-frozen and stored at —70°C pending testing.
Representative samples were tested for sterility and virus
concentrations. The material passing the in-process
quality control tests was thawed, centrifuged and the
supernatant pooled and distributed into ampoules. The
material was shell-frozen, kept at —70°C and finally
subjected to desiccation according to the method of
. The virus titrations at different production
2
Ward’
50 mouse
stages were performed using a standard LD
assay described elsewhere’’.
Stability of yellow fever vaccine: D.K. Sood et af.
Appropriate time for adding the stabilizer
Table 2
The four different stabilizers used consisted of: lactose,
sodium glutamate and gelatin (S
); medium 199, sodium
1
glutamate and gelatin (S
); lactose, D-SOrbitol, L
2
histidine hydrochioride and L-alanine hydrochioride
); sucrose and gelatin (S
3
(S
). All these stabilizers were
4
prepared in phosphate-bulTered saline with and without
calcium and magnesium ions. The control group was
prepared in double distilled water alone. Following
inoculation with seed virus, the eggs with living embryos
were harvested and divided into five groups containing
30—40 embryos per group. Chilled stabilizer solutions
and distilled water were added to half of the embryos of
the respective groups at the time of harvesting and to the
other portion at the time of filling. The vaccine lots were
lyophilized and titrated by mouse LD
50 assay.
Infectivity titres before and after lyophilization
Titre (mouse LD
)
1
50 mF
Reference
lot code
Before
After
Loss
A
B
C
0
E
F
G
H
5.60
4.90
4.90
5.10
5.10
5.05
5.05
4.60
5.00
4.34
4.34
4.51
4.74
4.41
4.38
4.50
0.60
0.56
0.56
0.59
0.36
0.64
0.67
0.10
Mean
S.d.
CV.
5.04
0.26
5.15
4.53
0.21
4.63
0.51
S.d., standard deviation; CV., coefficient of variation
Efficacy of ditferent stabilizers
The vaccine lots prepared with the different stabilizers
along with that prepared using distilled water, were tested
by the accelerated stability test after storing at 20°C
and exposure to 37°C for 7, 14, 21 and 28 days. The
resuits thus obtained were evaluated statistically.
0
—
5.0
Reproducibility in production metbodotogy
Three more batches of yellow fever vaccine were
prepared using stabilizer S
. The stabilizer was added at
3
the time of homogenization. These batches were tested
for accelerated stability by exposing to 37°C for 28 days
as described above.
/4.0’
9
o
-J
UJ
(1)
3,0
RESULTS
6
w
Virus infectivity titre after homogenization and at the
time of uiting
The results of virus infectivity titrations performed on
eight different lots after homogenization and during
fihling are shown in Table 1. The mean virus titre between
the two stages of production showed a difference of 0.43
(range 0.10—0.95) 1og
10 ml.
Infectivity titre before and after lyophilization
Table 2 shows that standard deviation and coefficient
of variation of eight different lots before and after
lyophilization were relatively constant. However, a loss
of 0.51 (range 0.10 to 0.67) 1og
10 per ml was observed
during desiccation.
Table 1 Comparative titrations of the intectivity titres at the
homogenization and at the time of tilling
Titre (mouse LD
0 mF’)
Reference
lot code
After
riomogenization
During
filling
Dit!erence
in titre
A
6
C
E
F
G
H
5.70
6.00
5.30
5.69
5.83
6.00
6.00
5.60
4.90
4.90
5.10
5.05
5.05
4.60
0.10
0.10
0.40
0.15
0.78
0.95
0.40
Mean value
5.72
5.03
0.43
Range
(5.25—6.00)
(4.60—5.60)
(0.10—0.95)
1-
2.0
1.0 L_
A
9
C
STQRAI3E CONDITIONS
D
Flgure 1 Infectivity titres of different batches of yellow lever vaccine af
different temperatures. Storage conditions: A, —20°C (original titre); B,
—20°C for 1 year; C, +4°C for 1 year; 0, +37°C for 14 days
Virus titre of the vaccine exposed to dilterent temperatures
The eight lots of yellow fever vaccine belonging to 42
different batches were titrated for virus infectivity after
storage at —20°C and 4°C for 1 year and 37°C for 14
days. The resuits so obtained were compared with the
original titre (Figure 1). All the samples stored at —20°C
and 4°C for 1 year retained a minimum titre of 1og
10 3.0
per human dose. There was a statistically significant
difference between the original virus titre and those kept
at 37°C and 4°C for 1 year (p <0.01).
Thermostability behaviour of the vaccine prepared
without stabilizers
Mean loss in virus titre among different storage
conditions with respect to original titre is shown in
Vaccine, Vol. 11, Issue 11, 1993
1125
Stability of yellow foyer vaccine: D.K. Sood et al.
Figure 2. All the batches stored at —20°C for a year had
a loss in virus titre of <1.0 1og
10 while the same was
true of 70% of those kept at 4°C for 1 ycar and 25% of
those exposed to 37°C for 14 days. Analysis of variance
showed that thermodegradation after storage al —20°C
and 4°Cfor 1 yeardidnotdiffersignificantly(p >0.01).
titre per ml at different stages of vaccine preparation is
shown in Table 3. The loss in virus titre was statistically
less (p <0.05) when stabilizers were added at the time
of homogenization than before filling.
Efficacy of different stabilizers and reproducibility in the
production methodology
Appropriate time to add stabilizers
Eflicacy of difTerent stabilizers in terms of loss of virus
titre after exposure to 37°C for 7, 14, 21 and 28 days was
studied by performing virus titrations. All the stabilizers
worked well when compared with the aqueous-based
vaccine (Figure 3). The difference in the loss of virus titre
with respect to different stabilizers and period of exposure
to ambient temperature (Figure 4) was statistically
significant (p <0.01). Reproducibility in production
methodology using stabilizer S
3 was demonstrated by
Four different stabilizers, with distilled water as
control, were used to prepare the vaccine. Loss of virus
3.0
2.0
5.0
S
1.5.
4.0
1
CD
0
-J
0
UJ
cr
0
Lf,
t—
3.0
0
-J
1.0
t—
z
w
(n
requiremerit
0
-J
-J
2 .0
11
0.5’
1 .0
0.0
7days
-20’t
L
-2fiC/YE.AR •4c;/YEAR •37b/14DAYS
Figure 3
Efficacy of different stabilizers in an accelerated stability test.
*, S;
distilled water (control)
; l. S
2
;
3
0. S,; L, S
•,
Appropriate time of addition of stabilizers
Titre (mouse L0
50 mfl)’
Loss in virus titre/mI
A
B
C
A and C
Percentage loss
Added at the time of homogenization
S,
2
S
3
S
4
S
DW
4.97
4.92
5.02
5.02
4.96
4.36
4.70
4.70
4.40
4.36
4.40
4.54
4.55
3.92
3.72
0.57
0.38
0.47
1.03
1.24
11.47
7.72
9.36
20.51
25.00
Added at the time of filling
S,
2
S
3
S
4
S
0W
5.10
5.10
5.10
5.10
5.10
4.50
4.70
4.65
4.40
4.46
4.30
4.50
4.50
4.00
3.72
0.80
0.60
0.60
0.70
1.28
15.68
11.76
11.76
13.72
25.09
Experiment
2
28doys
ST0RAGE cONDIrIONs
Figure 2 Thermal degradation of yellow fever vaccine after storage at
different temperatures
1
2ldays
37C
STORAGE CONDITIONS
Table 3
I4doys
Stabilizers
0W, distilled water
2, Virus titre at the time of homogenization; B, virus titre at the time of filling; C, virus titre after lyophilization
A
1126
Vaccine, Vol. 11, Issue 11, 1993
Stability of yel!ow fever vaccine: 0K. Sood et al.
3.0
9
CD
0
-J
2.0
Although the vaccine is freeze-dried, lyophilization
provides no guarantee for the preservation of potency
because at a given temperature, the rate at which the
titre drops depends upon factors such as virus strain,
residual moisture content and stabilizer used, if any. It
is, therefore, necessary to check the potency of such a
freeze-dried preparation under different conditions. The
mean virus titre of 42 batches stored at 20°C and 4°C
for 1 year and 37°C for 14 days is depicted in Figure 1.
The further loss in virus titre under different storage
conditions with respect to the original titre (Figure 2)
shows that all the batches stored at —20°C for 1 year
had a loss in virus titre of 0.1 log,
0 while the loss was
70% in those kept at 4°C for 1 year and only 25% in
those cases exposed to 37°C for 14 days. This is in
accordance with the findings of other 33
workers
.
From the data it is dear that at present yellow fever
vaccine produced at Central Research Institute, Kasauli,
which follows all the quality control procedures, stili
requires the addition of stabilizers to comply with the
new WHO regulation on stability criteria
.
4
The basic problem in dealing with preservation of viral
vaccines is the loss in virus titre either during
lyophilization and/or due to improper cold chain
conditions. As far as the first case is concerned, when the
frozen material is dried by selective sublimation, the virus
has to be protected from sudden physicochemical stress
by adding some cryoprotective agents to cover the
essential infective components of the nucleic acid’
. In
2
the second case, a number of sugar solutions along with
amino acids have helped in stabilizing the different
freeze-dried viral preparations subjected to ambient
temperature’ Therefore, to increase the stability
.
4
behaviour of conventional yellow fever vaccine, four
different stabilizers were tried. On statistical evaluation,
the optimum time for adding the stabilizers was found
to be at the time of homogenization of infected embryos
(Table 3). Tannock ei al.
2 were also able to produce a
more stable vaccine by homogenizing the embryos in
protein-free stabilizers.
To evaluate the efficacy of different stabilizers, four
different preparations of the vaccine were stored at 37°C
for 4 weeks and titrations were carried Out 0fl each group
at weekly intervals keeping a vaccine with an aqueous
base as control (Figure 3). Stabilizers S,, S
2 and S
3 kept
the vaccine stable for up to 4 weeks, but S
4 for only 3
weeks. However, the stability behaviours of all four
preparations fulfilled the stability criteria of WHO
.
4
The loss in virus titre (K value) in different vaccine
preparations after exposure to 37°C for 4 weeks with
respect to one kept at —20°C (Figure 2) showed that
statistically there was a significant difference between the
effect of different stabilizers (p < 0.01). Statistically,
stabilizer S
3 was found to be better than S, or S
, similar
2
to the findings of Barme and 5
Bronnert Using the same
.
stabilizer, a loss of 0.35 p.f.u. per dose was reported when
the vaccine so prepared was exposed to 37°C for 14 days.
Further reproducibility in the production methodology
was established by preparing and testing small batches
of the vaccine with S
3 stabilizers and subjecting them to
an accelerated stability test (Table 3). All the batches
met the stability requirement of yellow fever vaccine.
Barme ei al.
6 also prepared five lots of the vaccine using
the same stabilizer. These batches were found to be stable
when tested at 37°C for 14 days, 22°C for 6 months and
4°C for 3 years.
—
1—
z
-J
-J
‘t .0
0.0
1
3
2
4
STORAGE AT 37•C (WEEKS)
Figure 4 Evaluation of different stabilizers with respect to loss of virus
titre. Symbots as in Figure 3
Table 4 Reproducibility in methodotogy of production of stabilized
yellow fever vaccine
Titre (mouse 0
LD
/
dose)
37°C (days)
Batch no.
—20°C
7
14
21
28
II
III
4.50
4.06
4.10
4.06
3.77
3.95
3.86
3.68
3.83
3.75
3.60
3.73
3.65
3.61
3.69
Mean of three ampoules tested for each exposure
preparing and testing three batches of the vaccine
(Table 4).
DISCUSSION
Central Research Institute, Kasauli is one of the twelve
WHO-approved centres for the production and supply
of vaccine for international travellers. The summary
protocols of the WHO
11 on the finished produets of each
lot were sent to the International Quality Control
Division on Biologicals for scrutiny and final approval
for use in humans. As the production protocols have
been retained by the manufacturer, very little information
is available on the detailed production procedures of the
vaccin& In the first part of the study an attempt has
.
3
been made to evaluate the parameters for controlling the
production methodology of the yellow fever vaccine.
Virus infectivity titres of the embryos after homogen
ization and at the time of filling indicate that one
freeze-thaw cycle needed during production of the vaccine
has no adverse effect on vjrus infectivity (Table 1). To
assess the efficacy of the lyophilization process further,
a loss of 0.51 log,
0 ml of virus infectivity was noted
during desiccation of the liquid vaccine (Table 2). These
findings are in accordance with the data published by
Penna’° and De’Souza Lopes ei al.’
. This clearly shows
3
that the production procedures at Central Research
Institute, Kasauli meet the international standards.
Vaccine, Vol. 11, lssue 11,
1993
1127
Stability of yellow fever vaccine: D.K. Sood et al.
The vaccine so prepared not only fulfils all the
requirements of WHO (1976)11 but has also been found
to be safe and efficacious in a controlled field trial.
Seroconversion studies in a few hurnans have indicated
that the stabilized vaccine invokes immunity. It can only
be ascertained how long this immunity lasts after
follow-up studies of the immunized population.
4
5
6
7
ACKNOWLEDGEM ENTS
The technical assistance of the staff of the Yellow Fever
Vaccine Section of the Central Research Institute,
Kasauli and the secretarial assistance of Mr Raj Kumar
in typing this manuscript are gratefully acknowledged.
8
9
10
REFERENCES
11
1
2
Ç
3
Smith, H.H., Penna, H.A. and Paoliello, A. Yellow lever vaccination
with culture virus (170) without immune serum. Am. J. Trop. Med.
Hyg. 1938, 18, 437—468
Tannock, GA., Wark, M.C. and Hair, C.C. The development of an
improved experimental yellow fever vaccine. J. Bio!. Stand. 1980,
8, 23—34
Update on new or improved vaccine. WHO/EPI/Gen/88, 1988, Vol. 3,
pp. 20—21
1128
Vaccine, Vol. 11, Issue 11, 1993
12
13
14
Expert Committee on Biological Standardization. Requirement for
stability behaviours of yellow fever vaccine. WHO Tech. Rep. Ser
1987, 771
Barme, M. and Bronnert, C. Thermostabilisation du vaccin
antiamaril 17D lyophilisé 1. Essai de substances protectrices
J. Bio!. Stand. 1984, 12, 435—444
Barme, M., Vacher, B., Ryhiner, M.L. and Chabannier. G.
Thermostabilisation du vaccin antiamaril 17D lyophilisê II
Lots-pilotes preparés dans les conditions d’une production
industrielle. J. Bio?. Stand. 1987, 15, 67—72
De’Souza Lopes, 0., De Almedia Guimaraes, S.S.D. and
De’Carvalho R. Studies on yellow fever vaccine. II. Stability of the
reconstituted product. J. Bio?. Stand. 1988, 16, 71-76
Georges, A.J., Tible, F., Meunier, D.M.Y., Gonaalez, J.P.,
Beraud, A,M. et al. Thermostability and efficacy in the held of a
new, stabilized yellow tevervirus vaccine. Vaccine 1985,3, 313—315
Roche, J.C., Jonan, A., Brison, B., Rodhain, R., Fritzeil, 6. and
Hannoun, C. Comparative clinical study of a now 17D thermostable
yetlow lever vaccine. Vaccine 1986, 4, 163—165
Penna, H.A. Production of 17D yellow lever vaccine. WHO Monogr.
Ser. 1956, 30, 67—90
Expert Committee on Biological Standardization. Requirement for
yellow lever vaccine. WHO. Tech. Rep. Ser. 1976, 594
Ward, T.C. Method of storage and preservation of animal viruses.
Methods Viro!. 1968, 4, 481-489
De’Souza Lopes, 0., De Almedia Guimaraes, S.S.D. and
De Carvalho, R. Studies on yellow fever vaccine. t. Quality control
parameters. J. Bio!. Stand. 1987, 15, 323—329
Duffy, J.l. Stabilizer for live viral vaccines. In: Vaccine Preparation
Technique. Noyes Data Corporation, Park Ridge, NJ, 1980, p. 375
4
4h