of March 24, 2018.This information is current as

despite Sharing the Target AntigenTumor Cells without Damaging the Brain Immunotherapy Can Reject Intracranial

D. Lichty, Jonathan L. Bramson and Yonghong WanByram W. Bridle, Jian Li, Shucui Jiang, Ruby Chang, Brian

http://www.jimmunol.org/content/184/8/4269doi: 10.4049/jimmunol.0901447March 2010;

2010; 184:4269-4275; Prepublished online 17J Immunol 

Referenceshttp://www.jimmunol.org/content/184/8/4269.full#ref-list-1

, 17 of which you can access for free at: cites 46 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2010 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on March 24, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on March 24, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Immunotherapy Can Reject Intracranial Tumor Cells withoutDamaging the Brain despite Sharing the Target Antigen

Byram W. Bridle,* Jian Li,* Shucui Jiang,† Ruby Chang,† Brian D. Lichty,*

Jonathan L. Bramson,* and Yonghong Wan*

Although vaccines targeting tissue differentiation Ags represent a promising strategy for cancer immunotherapy, the risk of trig-

gering autoimmune damage to normal tissues remains to be determined. Immunizing against a melanoma-associated Ag, dopa-

chrome tautomerase (DCT), which normal melanocytes and glial cells also express, allowed concurrent analysis of autoimmune

consequences in multiple tissues. We show that vaccination with recombinant adenovirus expressing DCT elicited a strong CTL

response in C57BL/6 mice, leading to protection against intracranial challenge with B16-F10 melanoma cells. Both histological

analysis and behavioral testing indicated that there was no evidence of neuropathology in vaccinated animals and long-term sur-

vivors. Although vitiligo or demyelination could be induced by additional stimuli (i.e., surgery or inflammation) in DCT-vaccinated

mice, it did not extend beyond the inflammatory area, suggesting that there is self-regulatory negative feedback in normal tissues.

These results demonstrate that it is possible to vaccinate against a tumor embedded in a vital organ that shares the target Ag. The

Journal of Immunology, 2010, 184: 4269–4275.

Antigen-specific, active immunotherapy offers a promisingapproach to target both primary and metastatic cancers ofvarious types including intracranial (i.c.) tumors (1, 2). In this

regard, finding Ags exclusively expressed by tumor cells and in-corporating them into therapeutic vaccines will not only increase thelikelihood of induction of anti-tumor immunity but also reduce thepossibility of autoimmune pathology (3). However, development ofvaccines directed at tumor-specific Ags is complicated by the fact thateach vaccine will likely need to be patient specific (4). As an alternatestrategy, efforts have been invested into developing vaccines to targettumor-associated Ags (TAAs) that are typically nonmutated proteinsoverexpressed in specific classes of tumors (5). Indeed, despite the lackof mutation in these proteins, many cancer patients often displayspontaneous reactivity to TAAs. This realization has led to the de-velopment of different strategies that can overcome self-tolerance andelicit an immune response against self-Ags.Whether pathological damage to normal tissues that express TAAs

is a requisiteoutcomeof successful anti-tumor immunotherapy remainscontroversial.Evidence fromanimal studieshas shown that vaccinationagainst melanoma-associated Ags (MAAs), such as tyrosinase-relatedprotein 1 or dopachrome tautomerase (DCT; also known as tyrosinase-related protein 2), is often associated with damage to normal melano-cytes, leading to autoimmune vitiligo (6–9). Coupling of anti-tumorimmunity and autoimmunity has also been observed in other tumor

models, and furthermore, autoimmune symptoms appear to be a posi-tive sign of clinical efficacy in cancer immunotherapy trials (10–13).These results led to the hypothesis that autoimmune destruction is anunavoidable or necessary consequence of effective anti-tumor immu-nity. However, although the damage to nonessential organs may beacceptable, this hypothesis imposes a conceptual challenge to cancervaccines targeting Ags that are also expressed by cells in vital organs.Weandothershaveshownthat tumorprotectioncanbeachievedin the

absence of autoimmune disease (14–17). Specifically, we demonstratedin a murine melanoma model that a high-magnitude CTL response in-ducedby recombinant adenovirus (Ad) expressinghumanDCT (hDCT)did not result in vitiligo unless there was damage (i.e., trauma or in-flammation) to the normal tissue (15). A similar observation was madeby Lang et al. (18) that large numbers of CTL specific for pancreatic Agdid not necessarily result in autoimmune diabetes, and overt autoim-mune disease occurred only after an inflammatory stimulus that upre-gulated MHC class I expression in the pancreas. These data identifya concurrent inflammatory event in the target organ as a key factor de-termining the development of an autoimmune disorder following vac-cination against self-tumor Ags.To directly address the question of whether an effective vaccine im-

poses a risk to a vital tissue, we extended our melanoma vaccine studyfrom the skin to the brain. Because both melanocytes and CNS tissuedifferentiate from the neural crest during embryonic development, theyshare MAAs, including DCT (19–21). This provides an ideal experi-mental setting to examine the relationshipbetweenanti-tumor immunityand autoimmune pathology upon i.c. inoculation with melanoma cells.Furthermore,melanoma commonlymetastasizes to the brain in patients(22), and manyMAAs are expressed on humanmalignant gliomas (19,20, 23); thus, a better understanding of melanoma Ag-based vaccinesmay have a specific implication in the treatment of brain cancers.

Materials and MethodsMice and tumor cells

Age-matched (8–10 wk old at study initiation) female C57BL/6 (H-2b) mice(Charles River Laboratories, Wilmington, MA) were housed in specific path-ogen-free conditions. Animal studies complied with Canadian Council onAnimal Care guidelines and were approved byMcMaster University’s Animal

*Department of Pathology and Molecular Medicine and †Neurosurgery Division,Department of Surgery, McMaster University, Hamilton, Ontario, Canada

Received for publication May 7, 2009. Accepted for publication February 15, 2010.

This work was supported by grants to Y.W. from the Canadian Institutes of HealthResearch (MOP-67066) and the Ontario Cancer Research Network.

Address correspondence and reprint requests to Dr. Yonghong Wan, Department ofPathology and Molecular Medicine, McMaster University, Room MDCL-5024, 1200Main Street West, Hamilton, Ontario L8N 3Z5, Canada. E-mail address: wanyong@mcmaster.ca

Abbreviations used in this paper: Ad, recombinant adenovirus; DCT, dopachrometautomerase; DNFB, 2,4-dinitro-1-fluorobenzene; hDCT, human dopachrome tauto-merase; i.c., intracranial; MAA, melanoma-associated Ag; TAA, tumor-associatedAg.

Copyright� 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0901447

by guest on March 24, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

ResearchEthics Board.Murinemelanoma (B16-F10) cellswere grown in F11-MEM containing 10% FBS, 2 mM L-glutamine, 1 mM sodium pyruvate andvitamin solution, 0.01 mM nonessential amino acids, 50 mM 2-ME, 100 U/mlpenicillin, and100mg/ml streptomycin (all fromInvitrogen,Grand Island,NY).

Recombinant adenovirus. Adenoviral vector construction has been describedpreviously (24). Ad-hDCT expresses the full-length hDCT, and Ad-BHG con-tains no transgene.Viruseswere propagated in 293 cells and purified on a cesiumchloride gradient.

PeptidesImmunodominant peptides from DCT that bind to H-2Kb [DCT180–188,SVYDFFVWL; shared by hDCT and mDCT (25)] and I-Ab [hDCT88–102,KFFHRTCKCTGNFA (26)] were synthesized by Biomer Technologies(Hayward, CA).

Vaccination protocol

Anesthetized mice were immunized by i.m. injection of 1 3 108 PFU ofAd-hDCT in 100 ml PBS (50 ml/hamstring). Control mice received Ad-BHG or PBS.

Stereotactic surgery

To establish brain tumors or local inflammation,mice received i.c. injections of13 103–106 B16-F10 cells or 10 mg LPS in 2 ml saline. Mice were placed ina stereotaxis (Xymotech Biosystems, Mt. Royal, Quebec, Canada), and underboth general and local anesthesia, an incision was made in the scalp to exposethe skull. A needle mounted on a 10-ml Hamilton syringe (Hamilton, Reno,NV) was positioned over the right hemisphere of the brain, 2.25 mm lateral toBregma. A small burr hole was drilled through the skull, and the bevel of theneedle inserted into the brain parenchyma to a depth of 3 mm. Cells or LPSwere injectedover a periodof1min. The needlewas left in place for 2min priorto withdrawal to minimize reflux along the injection tract. The scalp incisionwas closed with stainless steel clips that were removed 7–10 d later.

Abs

The following mAbs were used in flow cytometry assays: anti-CD16/CD32(clone2.4G2)toblockFcRs,anti-CD3(clone145-2C11),anti-CD4(cloneRM4-5), anti-CD8 (clone 53-6.7) for detecting cell surface markers, and anti–IFN-g(clone XMG1.2) (all from BD Biosciences, San Diego, CA) for intracellularstaining. Immunodepletion studieswere conductedwith themAbsGK1.5 (anti-CD4) and/or 2.43 (anti-CD8) from American Type Culture Collection (Mana-ssas, VA). Purified mAbs (250 mg in 500 ml saline) were injected i.p. 2 d apartand then twice aweek thereafter at amaintenancedoseof 200mg/treatment.Theefficiency of specific depletion of lymphocyte subsets was.98% as measuredby flow cytometry. For phenotyping skin-infiltrating T cells, the following ad-ditional reagents were used: PE-conjugated Kb-SVYDFFVWL tetramer (TheProtein Core, Baylor College of Medicine, Houston, TX), recombinant mouseE-selectin/CD62E Fc chimera (R&DSystems,Minneapolis,MN), biotinylatedpolyclonal Ab to human IgG (Alexis Biochemicals, Plymouth Meeting, PA),PE-Cy7–conjugated streptavidin (BD Biosciences), and anti-CD194/CCR4(clone 2G12; BioLegend, San Diego, CA).

Detection of DCT-specific T cell responses

Ag-specific T cell responses were quantified by flow cytometric analysis 7 dpostvaccination.BloodwascollectedfromtheperiorbitalsinusandRBCslysed.To study skin-infiltratingT cells, postmortemdamagedandnormal shaved skinsamples measuring 13 5 cm were excised from each mouse. Damaged skinhadbeen injectedwithCFA48hprior toharvest or taken froma surgical lesion.Skin was minced, suspended in HBSS containing 0.1% collagenase (In-vitrogen), incubated in a shaker at 37˚Cat 200 rpmfor 45min, and thenpressedthrough a nylon mesh filter with a 40-mm pore size to obtain a single-cellsuspension. Mononuclear cells were stimulated with peptides (1 mg/mlDCT180–188+20mg/mlhDCT88–102) in thepresenceofbrefeldinA(GolgiPlug;BDPharmingen [SanDiego,CA], 1mg/ml added after 2 h of incubation).After6 h, total incubation time cells were treated with anti-CD16/CD32 and surfacemarkers fluorescently labeled by addition of Abs. Cells were then per-meabilized and fixedwith Cytofix/Cytoperm (BDPharmingen) and stained forintracellular cytokines.Datawere acquiredusing aFACSCanto flowcytometerwith FACSDiva 5.0.2 software (BD Pharmingen) and analyzed with FlowJoMac Version 6.3.4 software (Tree Star, Ashland, OR).

Histological analyses

Brainswere fixed for 3 d in 10% formalin, transferred to 70%ethanol, paraffin-embedded, sectioned at a thickness of 10 mm, and stained with H&E (Sigma-Aldrich Canada, Oakville, Ontario, Canada). To stain myelin, sections (10–25mm thick) were dehydrated in graded alcohol and placed in 0.1% Luxol fastblue at 37˚C overnight. Sections were then differentiated in 0.05% LiCO3 for

4 min and then dehydrated in graded alcohol and xylene (all reagents fromSigma-Aldrich, St. Louis, MO). For CD8 detection, brains frozen in optimalcutting temperaturemediumwere sectioned at a thickness of 10mm,fixedwithacetone, treated with peroxidase, stained with rat anti-mouse CD8 (clone 53-6.7; BD Biosciences) rat on mouse HRP polymer kit (Biocare Medical, Con-cord, CA), and resolved using an 3-amino-9-ethylcarbazole substrate kit (In-vitrogen).Myelinwas quantified bymeasuring integrated density of Luxol fastblue in stained sections taken through the injection site using ImageJ software(version 1.42q, http://rsb.info.nih.gov.ij; W. Rasband, National Institutes ofHealth, Bethesda, MD) (27)

Vitiligo assessment

Micewere immunizedwithAd-hDCT,and14d later, their skinswere“painted”with 120ml 2,4-dinitro-1-fluorobenzene (DNFB, 0.2%diluted in acetone/oliveoil at a 4:1 ratio) (Sigma-Aldrich).Vitiligodevelopmentwasmonitoredweeklyfor 8 wk after the DNFB challenge.

Behavioral assessments

Tofurtherevaluate the long-termeffectofvaccination therapyonbrain functions,a setofpreliminarybehavioral assessmentswereperformed inmice that survived.90 d posttherapy and age-matched unvaccinatedmice, as described previously(28). Briefly, each individual mouse was placed in an empty cage where objectavoidance, olfactory, auditory, and visual cliff tests were conducted. Neurolog-ical reflexes including balance, eye blink, ear twitch, whisker orientation, andrighting were assessed (29, 30). Upon completion of all tests for all animalswithin the same cage, behavior, such as locomotion, interaction with littermate,and grooming, were noted (31). A general health assessment consisting of bodyweight, body temperature, and fur condition were also recorded at the end of thetest session.

Statistical analyses

GraphPadPrismversion4.00forWindows(GraphPadSoftware,SanDiego,CA)was used to graph data and for statistical analyses. If required, data were nor-malized by log transformation. T cell responses were analyzed by Student two-tailed t test, one- or two-way ANOVA. Differences between means were con-sidered significant atp# 0.05.MeansplusSEbars are shown.Survival datawereanalyzed using the Kaplan-Meier method and the log-rank test.

ResultsAd-hDCT immunization induced effective protection againsti.c. challenge with B16-F10 cells

Murine melanoma B16-F10 cells are known for their aggressivenessand we have previously demonstrated that immunization with Ad-hDCTprovides robust protection against s.c. and challengewith these

FIGURE 1. Immunization with Ad-hDCT provides protection against i.c.

tumor challenge. A, Gross pathological analysis of brains (dorsal view) from

C57BL/6 mice that were harvested 13 d after an i.c. challenge with 1 3 103

B16-F10 cells. Seven days prior to challenge,micewere injected i.m.withAd-

BHG (left panel) or Ad-hDCT (right panel). Scale bar, 1 mm. B, Summary of

animal survivaldata followingvaccinationwithAd-BHG(n=13) orAd-hDCT

(n = 22). Of 22 mice immunized with Ad-hDCT, 6 never grew tumors (27%).

4270 SEPARATION OF ANTI-TUMOR ACTIVITY AND AUTOIMMUNE TOXICITY

by guest on March 24, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

cells (15, 32). We found that i.c. injection of 1 3 103 B16-F10 cellsresulted in formation of lethal tumors within only 13 d of injection(Fig. 1A, left panel).However, immunizationwithAd-hDCT7d priorto challenge effectively prevented tumor growth (Fig. 1A, rightpanel), demonstrating the potency of Ad-hDCTas a potential vaccinefor brain cancers that express DCT. This efficacy observed at earlytimepoints translated into a significant extension of survival (Fig. 1B).Mice treated with PBS (data not shown) or a control Ad vector (Ad-BHG) had a median survival of 15 d posttumor challenge, whereasimmunizationwithAd-hDCTresulted in a significant survival benefitin all treated mice (median = 64 d; p , 0.0001). More importantly,27% (6 of 22) of vaccinated mice completely rejected implanted tu-mor cells, offering a unique model for the subsequent analysis ofthe acute and chronic impact of vaccine-induced autoimmunity ina vital organ.

Protection against i.c. B16-F10 tumors depended on CD8+

T cells

To evaluate the induction of Ag-specific T cells by Ad-hDCT, wedetermined the frequency of both CD8+ and CD4+ T cells that werecapable of producing IFN-g in response to restimulation with theircorresponding peptides. At 8 d postvaccination, .3% of blood-derived CD8+ T cells were producing IFN-g in response to theimmunodominant epitope DCT180–188, shared between mice andhumans (Fig. 2A). A significant level ofCD4+T cells (0.4%) specificfor hDCT88–102, a recently identified helper peptide (26), was alsoevident by IFN-g production (Fig. 2B). To determine their relativecontribution to tumor protection, immunized mice were depleted ofCD4+ and/or CD8+ T cells 2 d before i.c. challenge. Protection wascompletely abrogated by depletion of CD8+ but not CD4+ T cells,indicating that CD8+T cells were the primary effectors for i.c. tumorrejection (Fig. 2C). This notion is further supported by histologicalstaining that revealed more immune infiltrates at the tumor chal-lenge site in DCT-vaccinated animals compared with controls (Fig.2D, upper panels), and many of these infiltrates were CD8+ T cells(Fig. 2D, lower panels).

No evidence of acute neuropathology despite autoimmunetargeting of DCT

For evaluation of potential acute autoimmune or bystander damage,micewere immunizedwithAd-hDCTandchallenged i.c.7d laterwitha high dose of 13 106 B16-F10 cells to ensure survival of tumor cellsso malignant lesions could be identified. Rapid dissemination of themelanoma to the ventral surface and brain stemwas observed in only

4 d (Fig. 3A, left panel). However, immunization with Ad-hDCT 7 dprior to challenge prevented tumor dissemination (Fig. 3A, rightpanel), confirming the potency ofAd-hDCT. Four days after the high-

FIGURE 3. Protection against high-dose tumor challenge in the absence of

overt neuropathology. A, Ventral view of brains harvested 4 d after i.c. chal-

lenge with 13 106 B16-F10 cells. Seven days before implantation, mice were

injected i.m. with Ad-BHG (left panel) or Ad-hDCT (right panel). Scale bar,

1 mm. B, These brains were sectioned and stained with H&E (left panels,

original magnification 3100) and Luxol fast blue (right panels, original

magnification 3100; to detect myelin) in adjacent serial sections to assess

potential damage to normal neural tissue. Upper panels, A naive mouse that

never received an i.c. challenge. The sections were taken at the depth at which

tumors were implanted in other mice.Middle and lower panels, Ad-BHG and

Ad-hDCT-treated mice, respectively. These sections represent those in which

the main tumor mass was largest. Circles indicate the location of the tumor at

the implantation site (or the equivalent area in the naive control). Squares show

aneural fiber-rich area suitable for evaluatingmyelination.Areas encompassed

by the circles and squares are shown at highmagnification. Scale bars, 100mm.

FIGURE 2. Tumor protection is mediated by CD8+

T cells. To evaluate effector mechanisms, mice received

i.m. injections of Ad-BHG or Ad-hDCT. TheDCT180–188-

specific CD8+ (A) andDCT88–102-specific CD4+ (B) T cell

responses were measured by flow cytometric detection of

intracellular IFN-g following in vitro peptide restim-

ulation. Each bar represents five mice. C, Fourteen days

post-Ad injection, mice received i.c. injections of 13 103

B16-F10 cells. Immunized mice were depleted of CD4+

and/or CD8+ T cells by administration of Abs 2 d prior to

and the day of tumor implantation and twice per week

thereafter. Survival data are shown (n = 5/group). D, In-

flammatory infiltrates (upper panels, original magnifica-

tion 3100) and CD8+ T cells (lower panels, original

magnification 340) were examined by H&E and immu-

nohistochemical staining, respectively, in sham-treated

(left panels) and Ad-hDCT–immunized (right panels)

mice subsequently challenged i.c. with 13 104 B16-F10

cells. Scale bars, 100 mm.

The Journal of Immunology 4271

by guest on March 24, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

dose challenge, mice were euthanized, their brains harvested, andfrozensectionswere stained (Fig. 3B)withH&E(left panels) orLuxolfast blue to detect myelin (right panels). A normal brain from an age-matched mouse was included. Again, the potency of Ad-hDCT im-munization on inhibition of tumor growth within the brain paren-chyma was evident relative to sham-treated controls (Fig. 3B, leftmiddle and left lower panels; see circles). This acute phase of DCT-specific cytotoxicity in the braindidnot trigger anyobvious damage tonormal tissues. Although swelling of tumor-adjacent brain tissuescould be observed in some untreated mice (example in Fig. 3A, leftmiddle panelwithhigh-magnification inset), their neurons andmyelinsheaths remained intact (Fig. 3A, right middle panel with high-magnification inset).

Immunization with Ad-hDCT caused autoimmune pathology ininflamed normal tissues

Wepreviously reported that i.m. immunizationwithAd-hDCT did notcause melanocyte damage (vitiligo) unless there was trauma/in-flammation within the skin (15). Consistent with this earlier observa-tion, all Ad-hDCT–immunized mice in the current model developedsevere vitiligo along the scalp incision site, whereas this lesion wasabsent in all sham-treated mice (Fig. 4A). Depletion of CD8+ T cellsbefore the surgery completely abrogated vitiligo induction, whereasCD4 depletion had a significant but incomplete effect, suggesting thatCD8+ T cells are the primary effectors in this autoimmune destruction(Fig. 4B). Tetramer staining indicated that a substantial number ofDCT-specific CD8+ T cells were indeed recruited into damaged skin

and 50.4% (6 SE of 3.5; n = 3) of these expressed E-selectin ligandand/or CCR4 (Fig. 4C), a surface phenotype consistent with skin-homing T cells (33). Interestingly, in all cases, the vitiligo was limitedto the area immediately surrounding the incision. To further delineatethe relationship between skin damage and vaccine-induced vitiligo,a hapten DNFB was used to induce skin sensitization in a controlledfashion. As shown in Fig. 4D, DNFB alone did not damage melano-cytes (top panel), whereas the size of the vitiligo lesion directlycorrelated with the extent of the DNFB-induced skin inflammation inAd-hDCT–immunized mice (bottom panel). The results reinforce theidea that the number of autoreactive T cells did not limit the extent ofvitiligo, but their access to normal skin was apparently controlled bythe degree of local inflammation.To determine whether additional inflammatory stimulation causes

autoimmune pathology in the brain, LPS was injected i.c. 14 d aftertreatment with Ad-hDCTor PBS. Compared with sham injection, LPSelicited more obvious inflammation, although it was limited to the in-jection site (Fig. 5, upper panels) and resolved in 3 d (data not shown).ConsiderablymoreCD8+Tcellswere evident inAd-hDCT–vaccinatedmice (Fig. 5, middle panels), which correlated with localized de-myelination (Fig. 5, lowerpanels;mean integrateddensity ofLuxol fastblue stain = 2.761.4 SE, n= 5 forAd-hDCT+LPS versus 10.262.6,n=5 for PBS+LPS;p=0.027). Sham injection orLPSchallenge alonewas insufficient to induce tissue damage (Fig. 5, lower panels). Theseobservations confirm the concomitant requirement of both Ag-specificautoimmunity and local tissue inflammation for the induction of au-toimmune sequelae.

FIGURE 4. Autoimmunevitiligo is associatedwith

damaged skin, but neural tissue remains intact in long-

term survivors. A, Mice receiving Ad-BHG and i.c.

challengedwithB16-F10 cells never developedvitiligo

(top panel). Mice immunized with Ad-hDCT and

challenged with B16-F10 cells always developed se-

vere vitiligo limited to the scalp incision area (bottom

panel). B, Mice vaccinated with Ad-hDCT had CD8+

(left panel) or CD4+ (right panel) T cells depleted right

before i.c.melanoma challenge.C, Skin dissected from

Ad-hDCT–immunized mice 48 h after injection of

CFA was used to assess the phenotype of infiltrating,

DCT-specific T cells by flow cytometry following tet-

ramer and Ab staining. A typical contour plot of

E-selectin ligand and CCR4 expression on skin-

infiltratingKb-DCT180–188+Tcells is shown (left panel)

alongwith a paired sample stainedwith isotype control

Abs (right panel). D, Mice treated with Ad-BHG (up-

per panel) or Ad-hDCT (lower panel) had their backs

shaved and the hapten DNFB painted on their skin 7 d

later. Only Ad-hDCT–immunized mice developed

vitiligo that was limited to the area of inflammation.

4272 SEPARATION OF ANTI-TUMOR ACTIVITY AND AUTOIMMUNE TOXICITY

by guest on March 24, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Long-term survivors did not have altered behaviors or chronicneuropathology

Long-term survivors generated in our study provided an opportunity toanalyze potential neural damage beyond static pathology. Twenty-onemice that survived i.c. challenge with B16-F10 cells and 22 age-matched naive mice were enrolled in a behavior study to assessmultiple sensory motor skills. This analysis included tests that eval-uated olfactory, visual, and auditory functions, all of which reflect theintegrity of CNS functionality. As summarized in Table I, mice thatsurvived their i.c. tumor challenge did not show any defects in thetested aspects, compared with controls. Furthermore, brains harvestedfrom these mice did not show any evidence of tumor cells, suggestingthesewere completely eradicated instead of their growth simply beingsuppressed. Notably, there was no evidence of any neuropathology in

these survivors, despite immunization with Ad-hDCT and tumor de-struction in the brain (data not shown). Because DCT-immunizedmice challenged with LPS had evidence of acute demyelination atthe injection site, we repeated the study to determine whether therewere any signs of chronic damage. At 90 d post-LPS challenge,these mice passed all behavioral tests (Table I) and did not have anydemyelination (mean integrated density of Luxol fast blue stain =241.16 1.0 SE, n = 5 for naive controls versus 240.26 1.9, n = 5 forLPS challenge).

Ad-hDCT vaccination provided a potent therapeutic effect

We next extended the analysis to evaluate the therapeutic potentialof Ad-hDCT in mice bearing pre-existing tumors in the brain. Inthis regard, mice were injected i.c. with 13 103 B16-F10 cells andthen treated with Ad-hDCT 5 d later. Eleven days after treatment,the whole brains were harvested from both naive (tumor-free) andtreated (tumor-bearing) mice for quantitative analysis of efficacy.Tumor burden was determined by subtracting the mean weight oftumor-free brains from those in the tumor-bearing groups. Fig. 6Ashows that the mean tumor weight in Ad-hDCT–immunized micewas 4.7-fold less than that in mice treated with a control Ad vector(47.1 6 21.3 versus 9.9 6 2.8 mg; p , 0.05). Furthermore, micetreated 5 d postengraftment had a significantly longer mediansurvival time following Ad-hDCT treatment, compared withsham-treated controls (Fig. 6B). An additional study extended thisobservation of Ad-hDCT–mediated therapeutic efficacy by dem-onstrating that immunization of mice with 7-d-old brain tumorsincreased median survival to 27 d versus only 15 for sham-treatedcontrols (p = 0.0067) (Fig. 6C). As had been observed in theprophylactic challenge model, vitiligo was evident in all micetreated in the therapeutic setting (Fig. 6D). However, these lesionswere less severe than those observed in the prophylactic model(Fig. 4A), presumably due to altered intervals between the peak ofthe anti–self-immune response (10–14 d after immunization) andsurgery-induced skin damage (1–2 d after surgery). Specifically,immunization was carried out several days after tumor in-oculation, such that a significant amount of wound healing hadoccurred by the time the CTL response was fully activated.

DiscussionStrategies to eliminate cancer cells by breaking tolerance to self-Agshave raised an issue regarding the potential risk for collateral auto-immunedamage tonormal tissues (34–36).Usingamurinemelanomamodel, we and others have demonstrated that the development of ef-fective anti-tumor immunity indeed results in autoimmune vitiligo,especially when there is a concurrent inflammatory event in healthytissues that express Ags carried by the vaccine (8, 15, 37, 38). Re-sults from several clinical trials also indicate that development of

FIGURE 5. Inflammation in the brain is only associated with damage in

Ad-hDCT-immunizedmice.Micewerevaccinatedwith PBS (middle panels)

or Ad-hDCT (right panels) for 14 d and then challenged i.c. with 10mg LPS.

Naive mice with sham injection (left panels) were included as controls. In-

flammatory infiltrates (top panels), including CD8+ T cells (middle panels),

were evaluated at the injection site by H&E and immunohistochemical

staining, respectively. Myelination at the injection site and surrounding area

was examined by Luxol fast blue staining and quantified as mean integrated

density6 SE (2.76 1.4, n = 4 for Ad-hDCT + LPS versus 10.26 2.6, n = 9

for PBS + LPS; p = 0.027; 4.3 6 2.2, n = 3 for sham injection). Scale bars,

100 mm (upper and middle panels, original magnification 340) or 500 mm

(lower panels, original magnification3100).

Table I. Mice protected from i.c. melanoma exhibit normal behavior

Mice

Physical Evaluation Neurological Reflex

ObjectAvoidance

AuditoryTest

Visual Cliff Test(Head Dip/min)a

Olfactory Test TimeSniffing (s)a Balance

WhiskerOrientation Righting Eye Blink Ear Twitch

Rejected tumors 21/21 21/21 11.88 6 1.30 5.53 6 0.99 20/21b 19/21c 20/21b 21/21 21/21Naive 27/27 27/27 11.02 6 0.82 5.73 6 2.50 27/27 27/27 27/27 27/27 27/27Ad-hDCT + LPS 5/5 5/5 10.40 6 2.48 4.59 6 0.79 5/5 5/5 5/5 5/5 5/5

Mice vaccinated with Ad-hDCT were challenged i.c. with B16-F10 cells or LPS 7–14 d later. Some mice completely rejected the implanted tumor cells. Twenty-one long-term survivors of the tumor challenge (.90-d survival), 5 mice at 90 d post-LPS challenge, and 27 age-matched naive controls were enrolled in a series of behavior studiesdesigned to assess general health and neurophysiological functions. Absence of brain tumors in these mice was confirmed by postmortem gross and histological examinationfollowing testing.

aMeans (6 SEM) were not significantly different.bThe mouse that failed these tests was the longest-term survivor and severely overweight.cThe whiskers on two of these mice were absent because of barbering.

The Journal of Immunology 4273

by guest on March 24, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

autoimmune toxicities correlated with tumor regression (13, 39–41).These observations point to the possibility that autoimmune pathol-ogy may be an unavoidable or even favorable outcome of cancerimmunotherapies.However, because the potential destruction of normal brain tissue

could have severe consequences, the balance between anti-tumor im-munity and autoimmune pathology in the case of cerebral malig-nancies needed to be investigated. The tumor model and vaccinationapproach presented in this study provided several unique features toaddress this issue. First, the potency of theAd-hDCTvaccine allowedus to achieve treatment efficacy against i.c. growth of B16-F10 mel-anomas and provided an opportunity to analyze pathological con-sequences both in early stages and in long-term survivors. Second, thepotentialautoimmunedestructioncouldbeconcomitantlyevaluatedinthe skin and brain, both of which express the target Ag. Third, i.c.challengewithB16-F10cellsdidnotonlyprovideabrain tumormodelbut also served as a local inflammatory stimulus within the brain.We previously reported that vitiligo lesions following melanoma

immunotherapyrequiredtwoevents:Ag-specificimmunityplusdamageto the skin (15). Thecurrent studycorroborates this. Specifically, vitiligowas associated with the scalpel incision site in Ad-hDCT–immunizedmice. Moreover, vitiligo was severe in the prophylactic model whereimmune responses were peaking at the time of injury to the skin. In

contrast, vitiligo severity was reduced in the therapeutic model where

.1 wk of wound healing occurred before significant DCT-specific

T cell responses were detectable. These observations reinforce our hy-

pothesis that autoimmune pathology following effective cancer vacci-

nation is the result of secondary trauma tonormal tissues, and its severity

is affected by the degree and persistence of normal tissue damage. We

speculate that tissue inflammation/trauma is required to recruit autor-

eactive T cells, as demonstrated in Fig. 4C, and to modulate expression

of MHC molecules and target Ags. Our results are consistent with the

finding by Lang et al. (18) that highly activated CD8+ T cells could

coexist with pancreaticb islet cells expressing the relevant autoantigen;

however, overt autoimmune disease occurred when an inflammatory

response was coupled.Despite the presence of robust DCT-specific CD8+ T cells and

the evidence that DCT is expressed in normal brain (19–21, 23,

42), we never saw any neuropathology in Ad-hDCT–immunized

mice. This may be explained by the inherent immune-privileged

nature of the brain (43) or the lack of inflammatory signals to re-

cruit autoreactive T cells into sites distal to the i.c. tumors. Our re-

sults confirm that Ag-specific vaccines are capable of inducing an

anti-tumor response within the immunologically privileged brain

and that the CNS is accessible to systemic immunotherapy (44).

More importantly, similar to what we observed in skin, immune

destruction appeared to only focus on the inflammatory area (i.e.,

tumor) and did not extend beyond the site of tumor growth. Fur-

thermore, the lack of behavioral anomalies or demyelination in

normal brain tissue by long-term survivors provides further evi-

dence that cerebral cancer can be eliminated by autoreactive

T cells while leaving the CNS unharmed. The linkage between

tissue inflammation and autoimmune pathology is further sup-

ported by deliberate i.c. injection of LPS where only Ad-hDCT–

immunized animals exhibit localized, acute demyelination, pre-

sumably due to the recruitment of CD8+ T cells that recognize

DCT expressed by normal glial cells. The fact that demyelination

was no longer evident at 90 d post-LPS challenge suggests the

cells that were acutely destroyed could be renewed following

cessation of inflammatory stimulation. Taken together, these re-

sults clearly demonstrate that the inflammation is tightly regulated

in normal tissues, which limits the initiation and progression of

autoimmune destruction in the vaccinated host.It is possible that additional stringent mechanisms may also be

operational in the brain to regulate local immune responses to avoid

autoimmune pathology. For instance, although most tumors are in-

herently inflammatory and the integrity of the blood-brain barrier is

locally compromised allowing extravasation of activated T cells,

their trafficking may be more limited in normal parenchyma (45).

However, it remains to be determinedwhether the local environment

functionally influences brain-infiltrating lymphocytes. The fact that

immunotherapy is less efficacious in i.c. versus s.c. compartments

using the same vaccination strategy suggests anatomical location of

the tumor can influence the outcome of cancer vaccines, with the

brain being a particularly difficult site (46).Overall, our findings provide proof-of-principle that self-Ags can

safely be targeted by cancer immunotherapy, even if the tumor is em-beddedwithin a vital tissue sharing the sameAg. Compelling evidencesupports the concept that anti-tumor immunity will only spill over intoautoimmune pathology if linked by inflammation. Given the demon-stration of MAA expression by gliomas, our findings on vaccinationefficacy and the relationship between anti-tumor immunity and auto-immune pathology have significant implications for the design oftherapies directed not only against brain melanoma but also potentiallyfor primary brain cancers.

AcknowledgmentsWe thank Duncan Chong, Xueya Feng, Ying Chuyan, Mary-Jo Smith, Mary

Bruni, Cai Jiang, and Jie Huang for technical assistance and Dr. Zhou Xing

for helping with histological examination.

DisclosuresThe authors have no financial conflicts of interest.

FIGURE 6. Therapeutic vaccination against established tumors. A, Mice

with 5-d-old brain tumors were treated with Ad-BHG (n = 4) or Ad-hDCT

(n = 5). Ten days later, Ad-treated mice and 10 age-matched, tumor-free

controls were euthanized, and their brains were weighed. The mean weight

of control brains was subtracted from those in the tumor-bearing groups to

quantify tumor burden. Mean tumor weights are shown with SE bars. Mice

carrying 5-d (B) or 7-d-old (C) tumors were treated with 1 3 108 PFU of

Ad-hDCT or Ad-BHG. Survival data are shown. Treatment of 5-d-old

tumors has been repeated 10 times with similar results. D, Vitiligo was

manifested in all Ad-hDCT–treated animals, which was limited to the

scalp incision area.

4274 SEPARATION OF ANTI-TUMOR ACTIVITY AND AUTOIMMUNE TOXICITY

by guest on March 24, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

References1. Mitchell, D. A., P. E. Fecci, and J. H. Sampson. 2008. Immunotherapy of ma-

lignant brain tumors. Immunol. Rev. 222: 70–100.2. Blattman, J. N., and P. D. Greenberg. 2004. Cancer immunotherapy: a treatment

for the masses. Science 305: 200–205.3. Schietinger, A., M. Philip, and H. Schreiber. 2008. Specificity in cancer immuno-

therapy. Semin. Immunol. 20: 276–285.4. Mufson, R. A. 2006. Tumor antigen targets and tumor immunotherapy. Front. Biosci.

11: 337–343.5. Malyankar, U. M. 2007. Tumor-associated antigens and biomarkers in cancer and

immune therapy. Int. Rev. Immunol. 26: 223–247.6. Engelhard, V. H., T. N. Bullock, T. A. Colella, S. L. Sheasley, and D. W. Mullins.

2002. Antigens derived from melanocyte differentiation proteins: self-tolerance,autoimmunity, and use for cancer immunotherapy. Immunol. Rev. 188: 136–146.

7. Overwijk, W.W., D. S. Lee, D. R. Surman, K. R. Irvine, C. E. Touloukian, C. C. Chan,M. W. Carroll, B. Moss, S. A. Rosenberg, and N. P. Restifo. 1999. Vaccination witha recombinant vaccinia virus encoding a “self” antigen induces autoimmune vitiligoand tumor cell destruction in mice: requirement for CD4+ T lymphocytes. Proc. Natl.Acad. Sci. USA 96: 2982–2987.

8. Overwijk, W. W., M. R. Theoret, S. E. Finkelstein, D. R. Surman, L. A. de Jong,F. A. Vyth-Dreese, T. A. Dellemijn, P. A. Antony, P. J. Spiess, D. C. Palmer, et al.2003. Tumor regression and autoimmunity after reversal of a functionally tol-erant state of self-reactive CD8+ T cells. J. Exp. Med. 198: 569–580.

9. Steitz, J., J. Bruck, J. Lenz, S. Buchs, and T. Tuting. 2005. Peripheral CD8+

T cell tolerance against melanocytic self-antigens in the skin is regulated in twosteps by CD4+ T cells and local inflammation: implications for the pathophys-iology of vitiligo. J. Invest. Dermatol. 124: 144–150.

10. Chianese-Bullock, K. A., E. M. Woodson, H. Tao, S. A. Boerner, M. Smolkin,W. W. Grosh, P. Y. Neese, P. Merrill, G. R. Petroni, and C. L. Slingluff, Jr. 2005. Auto-immune toxicities associatedwith theadministrationof anti-tumorvaccinesand low-doseinterleukin-2. J. Immunother. 28: 412–419.

11. Luiten, R. M., E. W. Kueter, W. Mooi, M. P. Gallee, E. M. Rankin, W. R. Gerritsen,S. M. Clift, W. J. Nooijen, P. Weder, W. F. van de Kasteele, et al. 2005. Immu-nogenicity, including vitiligo, and feasibility of vaccination with autologous GM-CSF–transduced tumor cells in metastatic melanoma patients. J. Clin. Oncol. 23:8978–8991.

12. Ram, M., and Y. Shoenfeld. 2007. Harnessing autoimmunity (vitiligo) to treat mel-anoma: a myth or reality? Ann. N. Y. Acad. Sci. 1110: 410–425.

13. Attia, P., G. Q. Phan, A. V. Maker, M. R. Robinson, M. M. Quezado, J. C. Yang,R. M. Sherry, S. L. Topalian, U. S. Kammula, R. E. Royal, et al. 2005. Autoim-munity correlates with tumor regression in patients with metastatic melanomatreated with anti-cytotoxic T-lymphocyte antigen-4. J. Clin. Oncol. 23: 6043–6053.

14. Turk, M. J., J. D.Wolchok, J. A. Guevara-Patino, S. M. Goldberg, and A. N. Houghton.2002. Multiple pathways to tumor immunity and concomitant autoimmunity. Immunol.Rev. 188: 122–135.

15. Lane, C., J. Leitch, X. Tan, J. Hadjati, J. L. Bramson, and Y. Wan. 2004. Vacci-nation-induced autoimmune vitiligo is a consequence of secondary trauma to theskin. Cancer Res. 64: 1509–1514.

16. Bronte, V., E. Apolloni, R. Ronca, P. Zamboni, W. W. Overwijk, D. R. Surman,N. P. Restifo, and P. Zanovello. 2000. Genetic vaccination with “self” tyrosinase-related protein 2 causes melanoma eradication but not vitiligo. Cancer Res. 60:253–258.

17. Hodge, J. W., D. W. Grosenbach, W. M. Aarts, D. J. Poole, and J. Schlom. 2003.Vaccine therapy of established tumors in the absence of autoimmunity. Clin.Cancer Res. 9: 1837–1849.

18. Lang, K. S., M. Recher, T. Junt, A. A. Navarini, N. L. Harris, S. Freigang, B. Odermatt,C. Conrad, L. M. Ittner, S. Bauer, et al. 2005. Toll-like receptor engagement convertsT-cell autoreactivity into overt autoimmune disease. Nat. Med. 11: 138–145.

19. Chi,D.D.,R.E.Merchant,R.Rand,A. J.Conrad,D.Garrison,R.Turner,D.L.Morton,and D. S. Hoon. 1997. Molecular detection of tumor-associated antigens shared byhuman cutaneous melanomas and gliomas. Am. J. Pathol. 150: 2143–2152.

20. Prins, R. M., S. K. Odesa, and L. M. Liau. 2003. Immunotherapeutic targeting ofshared melanoma-associated antigens in a murine glioma model. Cancer Res.63: 8487–8491.

21. Jiao, Z., Z. G. Zhang, T. J. Hornyak, A. Hozeska, R. L. Zhang, Y. Wang, L. Wang,C. Roberts, F. M. Strickland, and M. Chopp. 2006. Dopachrome tautomerase (Dct)regulates neural progenitor cell proliferation. Dev. Biol. 296: 396–408.

22. Madajewicz, S., C. Karakousis, C. R. West, J. Caracandas, and A. M. Avellanosa.1984. Malignant melanoma brain metastases: review of Roswell Park MemorialInstitute experience. Cancer 53: 2550–2552.

23. Liu, G., H. T. Khong, C. J. Wheeler, J. S. Yu, K. L. Black, and H. Ying. 2003.Molecular and functional analysis of tyrosinase-related protein (TRP)-2 as a cyto-toxic T lymphocyte target in patients with malignant glioma. J. Immunother. 26:301–312.

24. Yang, T. C., J. Millar, T. Groves, N. Grinshtein, R. Parsons, S. Takenaka, Y. Wan,and J. L. Bramson. 2006. The CD8+ T cell population elicited by recombinantadenovirus displays a novel partially exhausted phenotype associated with pro-longed antigen presentation that nonetheless provides long-term immunity.J. Immunol. 176: 200–210.

25. Parkhurst, M. R., E. B. Fitzgerald, S. Southwood, A. Sette, S. A. Rosenberg, andY. Kawakami. 1998. Identification of a shared HLA-Ap0201–restricted T-cellepitope from the melanoma antigen tyrosinase-related protein 2 (TRP2). CancerRes. 58: 4895–4901.

26. Kianizad, K., L. A. Marshall, N. Grinshtein, D. Bernard, R. Margl, S. Cheng,F. Beermann, Y. Wan, and J. Bramson. 2007. Elevated frequencies of self-reactiveCD8+ T cells following immunization with a xenoantigen are due to the presence ofa heteroclitic CD4+ T-cell helper epitope. Cancer Res. 67: 6459–6467.

27. Collins, T. J. 2007. ImageJ for microscopy. Biotechniques 43(Suppl. 1): S25–S30.

28. Karl, T., R. Pabst, and S. von Horsten. 2003. Behavioral phenotyping of mice inpharmacological and toxicological research. Exp. Toxicol. Pathol. 55: 69–83.

29. Crawley, J. N., and R. Paylor. 1997. A proposed test battery and constellations ofspecific behavioral paradigms to investigate the behavioral phenotypes of trans-genic and knockout mice. Horm. Behav. 31: 197–211.

30. Miyakawa, T., E. Yared, J. H. Pak, F. L. Huang, K. P. Huang, and J. N. Crawley.2001. Neurogranin null mutant mice display performance deficits on spatiallearning tasks with anxiety related components. Hippocampus 11: 763–775.

31. Walsh, R. N., and R. A. Cummins. 1976. The Open-Field Test: a critical review.Psychol. Bull. 83: 482–504.

32. Leitch, J., K. Fraser, C. Lane, K. Putzu, G. J. Adema, Q. J. Zhang, W. A. Jefferies,J. L. Bramson, and Y. Wan. 2004. CTL-dependent and -independent anti-tumorimmunity is determined by the tumor not the vaccine. J. Immunol. 172: 5200–5205.

33. Dudda, J. C., J. C. Simon, and S. Martin. 2004. Dendritic cell immunizationroute determines CD8+ T cell trafficking to inflamed skin: role for tissue mi-croenvironment and dendritic cells in establishment of T cell-homing subsets.J. Immunol. 172: 857–863.

34. Koon, H., and M. Atkins. 2006. Autoimmunity and immunotherapy for cancer.N. Engl. J. Med. 354: 758–760.

35. Gilboa, E. 2001. The risk of autoimmunity associated with tumor immunotherapy.Nat. Immunol. 2: 789–792.

36. Caspi, R. R. 2008. Immunotherapy of autoimmunity and cancer: the penalty forsuccess. Nat. Rev. Immunol. 8: 970–976.

37. Steitz, J., C. M. Britten, T. Wolfel, and T. Tuting. 2006. Effective induction of anti-melanoma immunity following genetic vaccination with synthetic mRNA codingfor the fusion protein EGFP.TRP2. Cancer Immunol. Immunother. 55: 246–253.

38. Uchi, H., R. Stan, M. J. Turk, M. E. Engelhorn, G. A. Rizzuto, S. M. Goldberg,J. D. Wolchok, and A. N. Houghton. 2006. Unraveling the complex relationship be-tween cancer immunity and autoimmunity: lessons from melanoma and vitiligo. Adv.Immunol. 90: 215–241.

39. Kaufman, H. L., and J. D. Wolchok. 2006. Is tumor immunity the same thing asautoimmunity? Implications for cancer immunotherapy. J. Clin. Oncol. 24:2230–2232.

40. Gogas, H., J. Ioannovich, U. Dafni, C. Stavropoulou-Giokas, K. Frangia, D. Tsoutsos,P. Panagiotou, A. Polyzos, O. Papadopoulos, A. Stratigos, et al. 2006. Prognostic sig-nificance of autoimmunity during treatment of melanoma with interferon. N. Engl.J. Med. 354: 709–718.

41. Dudley,M.E., J.R.Wunderlich,P.F.Robbins, J.C.Yang,P.Hwu,D. J.Schwartzentruber,S. L. Topalian, R. Sherry,N. P. Restifo, A.M.Hubicki, et al. 2002. Cancer regression andautoimmunity in patients after clonal repopulationwith anti-tumor lymphocytes. Science298: 850–854.

42. O, I., M. Blaszczyk-Thurin, C. T. Shen, and H. C. Ertl. 2003. A DNA vaccineexpressing tyrosinase-related protein-2 induces T-cell-mediated protectionagainst mouse glioblastoma. Cancer Gene Ther. 10: 678–688.

43. Mrass, P., and W. Weninger. 2006. Immune cell migration as a means to controlimmune privilege: lessons from the CNS and tumors. Immunol. Rev. 213: 195–212.

44. Walker, P. R., T. Calzascia, V. Schnuriger, N. Scamuffa, P. Saas, N. de Tribolet,and P. Y. Dietrich. 2000. The brain parenchyma is permissive for full anti-tumorCTL effector function, even in the absence of CD4 T cells. J. Immunol. 165:3128–3135.

45. Engelhardt, B., and R. M. Ransohoff. 2005. The ins and outs of T-lymphocytetrafficking to the CNS: anatomical sites and molecular mechanisms. TrendsImmunol. 26: 485–495.

46. Broder, H., A. Anderson, T. J. Kremen, S. K. Odesa, and L. M. Liau. 2003.MART-1 adenovirus-transduced dendritic cell immunization in a murine modelof metastatic central nervous system tumor. J. Neurooncol. 64: 21–30.

The Journal of Immunology 4275

by guest on March 24, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from