DTIC ADA430067: Venezuelan Equine Encephalitis Replicon Immunization Overcomes Intrinsic Tolerance and Elicits Effective Anti-Tumor Immunity to the 'Self' Tumor-Associated Antigen, neu in a Rat Mammary Tumor Model

17 Pages·1.8 MB·English

by  Defense Technical Information Center

BreastCancerResearchandTreatment 82: 169–183,2003.
©2003KluwerAcademicPublishers. PrintedintheNetherlands.
Report
Venezuelan equine encephalitis replicon immunization
overcomes intrinsic tolerance and elicits effective anti-tumor
immunity to the ‘self’ tumor-associated antigen, neu in a rat
mammary tumor model
Edward L. Nelson1, Darue Prieto2, Terri G. Alexander1, Peter Pushko3, Loreen A. Lofts3,
JonathanO. Rayner4, KurtI. Kamrud4, Bolyn Fralish4, andJonathan F. Smith4
1Department of Medicine, Division of Hematology/Oncology, University of California, Irvine, CA; 2SAIC-
Frederick, NCI-FCRC; 3USAMRIID, Fort Detrick, Frederick, MD; 4AlphaVax, Inc., Research Triangle Park,
NC,USA
Keywords:breastcancer,immunotherapy,neu,rattumormodel,repliconvector
Summary
Many tumor-associated antigens (TAAs) represent ‘self’ antigens and as such, are subject to the constraints of
immunologictolerance.Therearesignificantbarrierstoelicitinganti-tumorimmuneresponsesofsufficientmag-
nitude. We have taken advantageof a Venezuelanequine encephalitis-derivedalphavirusreplicon vector system
withdocumentedinvivotropismforimmunesystemdendriticcells. Wehaveovercometheintrinsictoleranceto
the‘self’TAAratneuandelicitedaneffectiveanti-tumorimmuneresponseusingthisalphavirusrepliconvector
systemandadesignedtargetantigeninarigorousratmammarytumormodel.Wehavedemonstratedthecapacity
togenerate50%protectionintumorchallengeexperiments(p=0.004)andwehaveconfirmedtheestablishment
ofimmunologicmemorybybothsecondtumorchallengeandWinnAssay(p=0.009).Minorantibodyresponses
wereidentifiedandsupportedtheestablishmentofThelpertype1(Th1)anti-tumorimmuneresponsesbyisotype.
Animals surviving in excess of 300 days with established effective anti-tumor immunity showed no signs of
autoimmunephenomena.TogethertheseexperimentssupporttheestablishmentofTlymphocytedependent,Th1-
biasedanti-tumorimmuneresponsestoanon-mutated‘self’TAAinanaggressivetumormodel.Importantly,this
tumormodelissubjecttotheconstraintsofimmunologictolerancepresentinanimalswithnormaldevelopmental,
temporal, andanatomicalexpressionof a non-mutatedTAA. These data supportthe continueddevelopmentand
potentialclinicalapplicationofthisalphaviralrepliconvectorsystemandtheuseofappropriatelydesignedtarget
antigensequencesforanti-tumorimmunotherapy.
Introduction cancer patients have been documented to have im-
muneresponsestobreastcancertumor-associatedan-
OneineightAmericanwomenwillbediagnosedwith tigens (TAAs) [2–8], but these immune responses
breast cancer [1]. Despite significant treatment ad- are generally of low magnitude and are clearly
vances, asubstantialpercentageofwomendiagnosed insufficient to establish or maintain control of pa-
with breast cancer will develop metastatic disease, tients’ tumors. Many TAAs have been charac-
often after many years, suggesting the presence of terized as ‘altered self’ [9] or are mal-expressed
micrometastatic disease after initial treatment. Novel ‘self’ molecules, which may account, in part, for
treatment methods directed at seeking out and elim- the difficulty encountered in attempts to elicit ro-
inating this persistent micrometastatic disease might bust antigen-specific, anti-tumor immune responses
have substantialclinical benefit. The immunesystem due to intrinsic tolerance to ‘self’. Recent prom-
is particularly well suited for this purpose. Breast ising studies in non-Hodgkins lymphoma and mela-
Report Documentation Page Form Approved
OMB No. 0704-0188
Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington
VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it

does not display a currently valid OMB control number.
1. REPORT DATE 2. REPORT TYPE 3. DATES COVERED
08 AUG 2003 N/A -
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
Venezuelan equine encephalitis replicon immunization overcomes
intrinsic tolerance and elicits effective anti-tumor immunity to the ’self’ 5b. GRANT NUMBER
tumor-associated antigen, neu in a rat mammary tumor model, Breast
Cancer Research and Treatment 82:169-183 5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S) 5d. PROJECT NUMBER
Nelson, EL Prieto, D Alexander, TG Pushko, P Lofts, LA Rayner, JO
5e. TASK NUMBER
Kamrud, KI Fralish, B Smith, JF
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION
United States Army Medical Research Institute of Infectious Diseases, REPORT NUMBER
Fort Detrick, MD
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)
11. SPONSOR/MONITOR’S REPORT
NUMBER(S)
12. DISTRIBUTION/AVAILABILITY STATEMENT
Approved for public release, distribution unlimited
13. SUPPLEMENTARY NOTES
14. ABSTRACT
Many tumor-associated antigens (TAAs) represent self antigens and as such, are subject to the constraints
of immunologic tolerance. There are significant barriers to eliciting anti-tumor immune responses of
sufficient magnitude. We have taken advantage of a Venezuelan equine encephalitis-derived alphavirus
replicon vector system with documented in vivo tropism for immune system dendritic cells. We have
overcome the intrinsic tolerance to the self TAA rat neu and elicited an effective anti-tumor immune
response using this alphavirus replicon vector system and a designed target antigen in a rigorous rat
mammary tumor model. We have demonstrated the capacity to generate 50% protection in tumor
challenge experiments (p=0.004) and we have confirmed the establishment of immunologic memory by
both second tumor challenge and Winn Assay (p=0.009). Minor antibody responses were identified and
supported the establishment of T helper type 1 (Th1) anti-tumor immune responses by isotype. Animals
surviving in excess of 300 days with established effective anti-tumor immunity showed no signs of
autoimmune phenomena. Together these experiments support the establishment of T lymphocyte
dependent, Th1-biased anti-tumor immune responses to a non-mutated self TAA in an aggressive tumor
model. Importantly, this tumor model is subject to the constraints of immunologic tolerance present in
animals with normal developmental, temporal, and anatomical expression of a non-mutated TAA. These
data support the continued development and potential clinical application of this alphaviral replicon vector
system and the use of appropriately designed target antigen sequences for anti-tumor immunotherapy.
15. SUBJECT TERMS
breast cancer, immunotherapy, neu, rat tumor model, replicon vector
16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF
ABSTRACT OF PAGES RESPONSIBLE PERSON

a. REPORT b. ABSTRACT c. THIS PAGE SAR 15
unclassified unclassified unclassified
Standard Form 298 (Rev. 8-98)
Prescribed by ANSI Std Z39-18
170 ELNelsonetal.
noma have demonstrated the capacity of immu- vivarium conditions. Water and rodent chow were
notherapeutic strategies to elicit TAA-specific im- provided ad libitum. The rat mammary tumor cell
mune responses that are associated with clinical line13762MATBIII(CRL-1666,ATCC,Manassas,
responses [10–13]. However, the clinical bene- VA) was obtained and cultured in vitro as recom-
fit of anti-tumor immunotherapy in other solid mended. Cells were harvested using Versene (Gibco
tumors such as breast cancer remains to be demon- Life technologies, Rockville, MD) and washed three
strated. times in phosphate-buffered saline prior to suspen-
Alphaviruses, such as Venezuelan equine en- sion in injection grade normal saline for inoculation
cephalitis (VEE), are positive strand ribonucleicacid into recipient animals. BHK cells (CCL-10, ATCC,
(RNA)virusesthathaveseveralcharacteristicswhich Manassas,VA)wereusedforproductionandtitration
are potentially advantageous for the derivation of ofVRPs. Allworkwasperformedunderanapproved
anti-tumorimmunotherapeutic/vaccinevectorsystems and active animal experimental protocol. All exper-
[14,15], including demonstrated tropism for a subset iments were performed with strict adherence to all
of immune system dendritic cells (DCs) [16]. Sev- institutional animal care and use guidelines. The ex-
eralstrategiesusingvectorsystemsderivedfromalpha pression level of rat neu and MHC class I molecules
viruses have been described, including viral replicon were routinely monitored by flow cytometry (FAC-
particles(VRPs)[17–30].VRPsaresingle-cyclevec- Scan, B.D. Biosciences, San Diego, CA) usingFITC
tors containing RNA replicons with an engineered labeledappropriateisotypecontrols,Ab-4(Oncogene
multiple cloning site in place of the viral structural Science, Boston,MA)andOX-18(B.D.Biosciences,
protein genes. Heterologous genes cloned into this SanDiego,CA)respectively,Figure1.
site are expressed from the 26S subgenomic RNA
promoter at very high levels. Replicon particles are
Target antigen sequence. Partial protein sequences
produced by providing the missing structural pro-
from HER2/neu were used to probe the proteindata-
tein genes in trans on two helper RNAs [31]. The
bases for regions of homology to known proteins.
applicationofVEEVRPsforimmunizationwithvari-
Regions with the least amount of homology with
ous infectious disease antigens (one of which is also
other normal proteins, including other members of
a TAA) has been reported [27, 32–41]. The po-
the epidermalgrowth factor receptor (EGFR) family,
tency of VEE VRPs in these studies suggests that
were selected for inclusion in the target antigen se-
VRPsmighthavesimilarefficacyintargetinga‘self’
quence. Regionsfromthe extra-cellulardomainwere
TAA.
not included to optimize the likelihood of intracellu-
We have used a rat mammary tumor model and
lar expression, MHC class I processing/presentation,
selected the neu molecule, homologue of human
HER2/neu, as a prototypical TAA to test our hy-
pothesis that the application of VEE derived VRP
immunotherapywillovercomeintrinsictoleranceand
elicit efficacious anti-tumor immunity. The selec-
tion of a rat model is complicated by the availabil-
ity of fewer immunologic reagents, but the normal
expression pattern of rat neu, in contrast to mice
transgenic for rat neu, provides for a model; (1)
that more closely matches the human clinical situ-
ation, (2) with normal intrinsic tolerance, and (3)
the potential to observe elicited autoimmune phe-
nomenon.
Materialsandmethods
Figure1. 13762MATBIIexpressionofMHCclassIandratneu:
Representative, non-gated,datafromroutinefluorescentcytometry
Animalsandcelllines. Six-toeight-week-oldFisher
monitoringofthe13762MATBIItumorline. MHCclassI,bold
344 female rats (NCI-FCRC, Frederick, MD) were
line –noshading; ratneu, bold line –gray shading; andisotype
obtained and housed in grouped cages under normal controlthinline–noshading.
VEErepliconelicitedanti-tumorimmunityinaratmammarytumormodel 171
Figure 2. Comparative predicted amino acid sequences of human and rat c-erb B2. Standard single letter amino acid notations are used
for the human sequence above and rat sequence below. Vertical bars=identical amino acids, colon=conservative substitutions, and
periods=semi-conservativesubstitutions.Underlinedregionsrepresentthoseselectedforinclusioninthetargetantigensequence.
and T helper type 1 (Th1) immune responses. The (Gibco Life technologies, Rockville, MD) accord-
exceptional degree of homology between HER2/neu ing to manufacturers instructions. Pfu thermostable
and rat neu allowed the selection and constructionof polymerase (Stratagene, LaJolla, CA) was used for
anentirelyhomologoustargetantigensequenceforrat all PCR reactions. The following primers were used
neu, Figure 2. The selected sequences were obtained to amplify a small fragment that includes the trans-
byrtPCR frommRNA isolatedfromthe 13762MAT membrane domain, + strand (bases 1855–1882) 5(cid:3)-
B III tumor cell line. Total RNA was isolated using CTCCTACATGCCCATCTGGAAGTACCC-3(cid:3) and−
(cid:3)
Trizol reagent and first strand cDNA synthesis was strand (bases 2093–2120) 5-TAACTCAGTTTCCT
(cid:3)
performedusingSuperscriptIIReverseTranscriptase GCAGCAGCCTACG-3.Thefollowingprimerswere
172 ELNelsonetal.
used to amplify a larger fragment derived from the cloningintotheVEE VRPproductionconstructsand
cytoplasmic domain, + strand (modified from bases forproductionofHistaggedtargetantigenprotein.
(cid:3)
3110–3142) 5-GATTCTTCTCTCCGGAGCCCTAC
CCCAGGCAC-3(cid:3)and−strand(bases3920–3950)5(cid:3)- VEE VRP Production. The targetantigensequence,
(cid:3)
CAGCAAGGAAAGGTTCCTCGGGGCAGGTTC-3. designed and constructed as above, and the influ-
The proximal fragment contains an intrinsic BspE1 enzaAhemagglutinin(H1)sequenceweresubcloned
site and the terminal fragment positive strand primer intotheVEErepliconplasmidpVR200thathasbeen
was modified by changing the 11th base from a C described previously [41]. This plasmid along with
to T creating a BspE I site (underlined) and in- the two other plasmids encoding the structural gene
serting a G between bases 16 and 17 in order to sequences, the split helper plasmid system, were lin-
maintain the reading frame. These two fragments earizedbyNotIdigestion.Invitrotranscriptionusing
were digested with BspE I (New England Biolabs, T7 RNA polymerase was used to generate capped
Beverly, MA) under standard conditions, purified by RNAs that were electroporated into BHK-21 cells
agarose gel electrophoresis, and ligated using Rapid for the production of the VRPs. VRPs were con-
Ligation Kit (Gene Choice, PGC Scientifics Corp. centratedfromculturesupernatantsviacentrifugation
Frederick, MD) as per manufacturers instructions. througha20%sucrosecushionpreparedinPhosphate
The resultant ligation product was used for PCR Buffered Saline (PBS). Infectious unit titers (IU/ml)
amplification of the target antigen sequence using were obtained by plating serial dilutions on BHK-
the following primers containing engineered Bgl I 21monolayerswithimmunofluorescentevaluationof
sites for directional cloning (underline), + strand 5(cid:3)- VEE non-structural gene products or heterologous
CCCATCGCCACCATGGCCTCCTGTGTGGATCT protein expression. Extensive safety testing was per-
GGATGAACGAGGC-3(cid:3) and − strand 5(cid:3)-ACGTGC formedpriortoreleaseofVEEVRPsforexperimental
CCTTAAGGCTCATACAGGTACATCCAGGCCTAG evaluation to document the absence of replication
(cid:3)
GTACTC-3. The resulting target antigen sequence, competentvirus.VRPswereresuspendedinPBSwith
encoding a heterologous target antigen protein con- 1% normal rat serum and frozen at −80◦C for ship-
taining 275 amino acids, incorporates a ‘start’ me- mentanddilutedinthissamebuffertoanappropriate
thionine (M) and alanine (A) in the second position concentrationforadministration.
preceding the sequence from the smaller fragment.
Thefusionofthepolypeptidesencodedbythesmaller Immunization and phlebotomy. Animals were im-
andlargerPCRfragmentsrequiredaconservedsubsti- munized with 200µl of solution containing the ap-
tutionof arginine(R)forlysine (K)andthe insertion propriate concentration of VEE VRPs administered
ofserine(S)atthefusionsiteinordertomaintainthe througha 27g needle. All injectionswere performed
readingframethroughtheBspEIsite intheresultant with minimal restraining of conscious animals. The
fusion protein, Figure 3. This amplified tumor anti- administrationsitewascleansedwith70%ethanoland
gen sequencewas sub-clonedinto a standardcloning allowed to dry prior to immunization. Subcutaneous
vectorforsequenceconfirmationandsubsequentsub- (SC)immunizationswerelocatedapproximately0.5–
1.0cmcephaladandlateraltothebaseofthetailonthe
contralateralsideto tumorinoculation.Intramuscular
(IM) administrations were located in the quadriceps,
rectus medius. Ten to 12 days after completion of
immunizationsequences, venousblood was obtained
via standard saphenousvein phlebotomy. Serum was
storedat−20◦Cpriortoanalysis.
His-tagged rNeu protein expression and purifica-
tion. A histidine tag was added to the C-terminus
of the rat neu target antigen coding sequence
Figure3. Aminoacid sequence ofthe designed tumor-associated by PCR amplification using the following forward
targetantigen.Standardsingleletteraminoacidnotationsareused.
and reverse primers engineered to contain EcoRV
Underlined amino acids note modifications/additions to the wild (cid:3)
and AscI sites, respectively (sites underlined), 5-
type sequences, see text for full description, italicized sequence
(cid:3)
representsthetransmembranedomain. CGGATATCATGGCCTCCTGTGTGGATCTG-3 and
VEErepliconelicitedanti-tumorimmunityinaratmammarytumormodel 173
(cid:3)
5-TTGGCGCGCCTCAATGGTGATGGTGATGGTG ofgoatanti-ratIgG1-AP,a1:100dilutionofgoatanti-
(cid:3)
TACAGGTACATCCAGGCCTA-3. The PCR ampli- ratIgG2a-AP,ora1:75dilutionofgoatanti-ratIgG2b-
fied product was digested with EcoRV and AscI and AP (Bethyl Laboratories, Montgomery, TX). Serum
ligated into a similarly digested alphaviral replicon fromeachanimalwastestedinduplicatewitheachof
vector(pERK)[39]. thesecondaryantibodies.Inaddition,wellsthatwere
BHK cells were electroporated with RNA gener- coatedwithdilutionsofpurifiedIgG1,IgG2aorIgG2b
atedfromthereplicon-rNeu/Hisconstructasdescribed were incubated with the respective secondary anti-
in [41]. Sixteenhourspost-electroporationcellswere body,conditionswereestablishedinwhichtherewas
washedwithPBSandlysedinNP-40lysisbuffer(1% nocross-reactivity.Plateswerewashed,developedand
NP40,50mMNaPipH7.4,0.3MNaCl,10mMDOC, readasdescribedabove.
20mM Imidazole and mixture of protease inhibitors
(Roche, Indianapolis, IN)). The lysates were cleared Tumor challenge. Animals were immunized at 3-
◦
bycentrifugationat3000RPMfor15minat4 Cand weekintervalsforatotalofthreeimmunizationswith
subsequently filtered through a 5µM Millex SV fil- the appropriate dose and routes of administration for
ter (Millipore, Billerica, MA). The rNeu/His protein eachcohort(n=6). Tumorcells, 1×105 viablecells
was purified from the clarified lysates using Ni-NTA (preparations were >95% viable for use), were ad-
SuperflowColumns(Qiagen,Valencia,CA)following ministered into the SC space on the flank, located
themanufacturer’sprocedure. 1cm cephalad and lateral to the base of the tail. Tu-
mors in control animals developed in 12–14 days.
ELISA. Nunc-ImmunoMaxiSorbplates(NalgeNunc Tumor volumes were assessed and calculated using
International, Rochester, NY) were coated overnight the formula, volume=0.4(ab2) where ‘a’ and ‘b’
◦
at 4 C with 75ng of rNeu/His protein/well diluted represent perpendicular axis measurements with ‘a’
in carbonate–bicarbonate buffer pH 9.6 (Sigma, St. representingthelongestaxisdimension[42]. Tumors
Louis, MO). The plates were then blocked with 3% thatexceeded10ccoranysignsofdistressintheani-
◦
BSA(Sigma)inPBSfor1hat30 Candthenwashed mals were indicationsfor euthanizationthat was per-
six times with PBS. Serial dilutions of sera from formed by CO inhalation. Repeat tumor challenge,
2
experimental animals, diluted in 1×PBS with 1% an equivalent number of cells, was administered to
BSA and 0.05% Tween 20 (Sigma), were plated in survivinganimalsonday160frominitialtumorchal-
triplicate (50µl/well) and incubated for 1h at 30◦C. lenge, and placed contralateral to the original tumor
PlateswerewashedsixtimeswithPBSandincubated challenge.
1h with a 1:2000 dilution of HRP, Goat Anti-Rat,
IGG (H+L) (KPL, Gaithersburg, MD). After wash- Winn assay. Animals (n=10) received three im-
ing as above, 100µl of peroxidase substrate (ABTS munizations as above with 1×107IU VRPs. Two
Microwell Peroxidase Substrate System, KPL) was weeks after completing the immunization series, an-
addedto each well andplateswere read atOD405 on imalswere euthanizedandspleenswereharvested. T
a VERSAmaxmicroplatereader(MolecularDevices, cell enriched splenocytes were obtained by dissoci-
Sunnyvale,CA). ation of the spleen and passage of cellular material
through steel mesh, erythrocyte lysis (ACK buffer,
Isotyping. Sera from experimentalanimals that had Biosource International, Camarillo, CA), and pas-
detectableanti-rNeutargetantigenantibodywereiso- sage overpreviouslypreparedautoclavednylonwool
typed for IgG1, IgG2a and IgG2b responses. Nunc- columns (7.5g of nylon wool (Robbins Scientific
ImmunoMaxiSorbplates (NalgeNunc International) Corp., Sunnyvale, CA) packed into the barrel of a
◦
were coated overnightat 4 C with 75ng of rNeu/His 60cc syringe). Columns were incubated with RPMI
◦
protein/well or with purified IgG1, IgG2a or IgG2b 1640supplementedwith10%FCSat37 Cfor45min
(Southern Biotech, Birmingham, Alabama) for stan- prior to loading of the washed splenocytes. After in-
◦
dard curves from 500ng/well to 3.9ng/well diluted cubationat37 Cfor30minthenon-adherentcellular
in carbonate–bicarbonatebuffer, pH 9.6. Plates were fraction was eluted off the column with 2×column
blocked and washed as described above and sub- volumesofRPMI-1640supplementedwith10%FCS,
sequently incubated with rat sera diluted from 1:40 collectedandre-suspendedinPBS.Appropriatenum-
◦
to 1:5120for1h at30 C. Theplateswere washedas bers of immune T cell enriched splenocytes were
aboveandthenincubatedwitheithera1:250dilution addedtoasuspensionof13762MATBIIImammary
174 ELNelsonetal.
Figure4. (A)Tumorchallengesurvival.AnimalsreceivingVRP-rNeuintramuscularimmunizationsaredepictedbythedashedlineandopen
squares ((cid:1)), VRP-rNeu SC immunizations the circle dotted line and open circles (◦). Control animals, receiving VRP-HA intramuscular
immunizations,irrelevantantigenforvectorandspecificitycontrol,aredepictedbythesquaredotedlineandopentriangles((cid:7)),receivingSC
immunizationsof106irradiatedtumorcellbythesolidlineandopendiamonds(♦),andanimalsthatdidnotreceiveimmunizationbythethin
lineandclosedsquares((cid:3)).Animalswereeuthanizedwhentumorsexceededavolumeof10,000mm3.(B)Individualrattumorvolumes.Each
individuallinerepresentsasingleanimalinthecohortofVRP-rNeuintramuscularimmunizedanimals.Time0equalstumorinoculation.
VEErepliconelicitedanti-tumorimmunityinaratmammarytumormodel 175
Figure5. Tumorre-challengesurvival.Survivalofimmunizedanimalsreceivingasecondtumorchallenge140daysaftertheinitialchallenge.
VRP-rNeuintramuscularroutearedepictedbythedashedlineandopensquares((cid:1)),SCimmunizationroutebythedottedlineandopencircles
(◦),or106irradiatedtumorcellsbythesolidlineandopendiamonds(♦).
Figure6. Winnassay. Survival ofnaïve animals fromtimeofinoculation oftumorandnylonwool, T-cellenriched immunesplenocytes.
Splenocytes isolatedfromanimalsreceivingVRP-rNeuintramuscularimmunizationsaredepictedbythedashedlineandopensquares((cid:1)),
VRP-rNeuSCimmunizationsbythecircledottedlineandopencircles(◦),VRP-HA,theirrelevantantigen,bythesquaredottedlineandopen
triangles((cid:7))Control,andnon-immunizedsplenocytesbythethinlineandclosedsquares((cid:3)).Animalswereeuthanizedwhentumorsexceeded
avolumeof10,000mm3.
176 ELNelsonetal.
tumorcells,inappropriateamountstoallowfor200µl Table1. Post-immunizationmeanantibodytiters
inoculation volumes, just prior to administration to
Immunization Meanantibody Rangeofantibody
non-immunizednaïveanimalsas describedabovefor
titer titers
tumorchallenges.
107VRP-rNeuIM 2793 1920–3840
107VRP-rNeuSC 1975 640–3840
Results 106VRP-rNeuIM 2326 1280–5120
106VRP-rNeuSC 988 120–2560
ImmunizationwithVRPvectorsencodingtheratneu 105VRP-rNeuIM 2610 640–5120
105VRP-rNeuSC 1568 480–3840
target antigen sequence resulted in 50% of animals
107VRP-HAIM 135 <80–640
being protected from tumor challenge, Figure 4. Im-
107VRP-HASC 145 <80–640
munizationwithVRPencodingtheirrelevantantigen
Irradiated13762 <80 all<80
influenzahemagglutinin(HA) orsham immunization
Non-immunized <80 all<80
with vehicle alone resulted in no protection from
tumor challenge. The survival benefit for VRP-rNeu
immunizedanimalsrelative to controlanimalsis sta-
tistically significant, p=0.004 (two-tailed Fisher’s of six animals despite no previous demonstration of
Exact Test). The results depicted in Figure 4(a) are protectionin challengeexperiments.Flowcytometric
representativeofthreeseparateexperimentsthatgave evaluation of the nylon wool T cell enriched popu-
similar results. Interestingly, all animals receiving lations obtained from the various treatment cohorts
SC and a proportion of animals receiving IM VRP- demonstratednosignificantdifferencesinproportions
rNeu immunizations developed tumors with 25–50% of NK, CD3+, CD4+, CD8+ and MHC class II+
of these tumors permanently regressing, Figure 4(b). cells(datanotshown).
Overt protection from tumor challenge was observed We evaluated serum that was obtained from an-
only with the IM route of administration. There ap- imals approximately 10 days after completing the
pearedtobeamodestdose–responserelationshipover immunizationsequenceforantibodyresponsesto the
the range from 105 to 106 to 107IU/immunization ratneutargetantigenbyELISA.TheIMroutesofad-
(datanotshown). ministrationresultedinhighermeanantibodytitersat
Immunologicalmemorywasevaluatedusingboth all doses, Table 1. Controlanimals (sham, VRP-HA,
repeat tumor challenge and Winn assay experiments. andirradiatedtumorimmunized)didnotelicitsignif-
Animalsthathadrejectedtheirtumorsweresubjected icant anti-rat neu target antigen immune responses.
to asecondtumorchallenge140daysaftertheinitial This data is representative of the antibody titers for
tumorchallenge.Acohortofnaïveanimalswassimi- allthreeexperimentsinwhichserawascollected.All
larly challenged with tumor as a positive controland positivesampleswereevaluatedforisotyperesponses.
demonstrated the expected tumor development. All Only IgG2a or IgG2b isotypes were identified in the
surviving animals previously immunized with VRP- anti-rat neu target antigen reactive components. No
rNeu were protected from a second tumor challenge IgG1 anti-rat neu antibodies were detected (data not
while the control animals developed progressing tu- shown). There was no correlation with antibodytiter
morswithoutregression,Figure5.Therewerenosur- andrejectionoftumorchallenge.
viving animals fromthe non-immunizedor VRP-HA
immunizedcohorts.Immunologicmemoryelicitedby
the immunization procedure alone was assessed by Discussion
the Winn assay. Nylon wool T cell enriched immune
splenocytespurifiedfromimmunizedratswereadmin- This series of experiments demonstrates that VEE
isteredtonaïveanimalsadmixedwithtumorat100:1 derivedVRPsare anefficientvectorsystemforover-
and demonstrated significant protection, p=0.009 coming the existing level of tolerance and elicit-
(two-tailedFisher’sExactTest), fromtumordevelop- ing anti-tumor immunity to a highly conserved TAA
ment relative to animals that received nylon wool T which is also a normally expressed ‘self’ protein.
cell enriched non-immune splenocytes admixed with Theseresultswereobtainedinarattumormodelthat
tumor at the same ratio, Figure 6. Nylonwool T cell has expressionof the TAA in an entirely physiologic
enrichedVRP-HAimmunesplenocytesprotectedtwo manner in contrast to the various rat neu transgenic