cytotoxicity of human endogenous retrovirus speciﬁc t cells … · cancer cells kiera rycaj1,2,3,...

Biology of Human Tumors

Cytotoxicity of Human Endogenous RetrovirusK–Specific T Cells toward Autologous OvarianCancer CellsKiera Rycaj1,2,3, Joshua B. Plummer1,3, Bingnan Yin1,3, Ming Li1,3,4, Jeremy Garza1,Laszlo Radvanyi3,5,6, Lois M. Ramondetta7, Kevin Lin1, Gary L. Johanning1,3,Dean G. Tang2,3, and Feng Wang-Johanning1,3,4,6

Abstract

Purpose: To determine whether HERV-K envelope (ENV) pro-tein could function as a tumor-associated antigen and elicitspecific T-cell responses against autologous ovarian cancer cells.

Experimental Design: The expression of HERV-K transcriptsand ENV protein, the presence of serum antibodies against HERV-K, reverse transcriptase (RT) activities, and cellular immuneresponses in primary ovarian cancer tissues and patient bloodsamples were analyzed and comparedwith samples frompatientswith benign ovarian diseases and normal female donors.

Results: Ovarian cancer cells in primary tumors and ascitesexpressed markers of cancer stem cells and markers of bothmesenchymal and epithelial cells. Expression ofHERV transcriptsand HERV-K ENV protein and reverse transcriptase activities werehigher in ovarian cancer compared with adjacent normal andbenign tissues. The ovarian cancer patient plasma also had high

reverse transcriptase activities and the ovarian cancer patient seracontained HERV-K immunoreactive antibodies. HERV-K–specificT cells generated from autologous dendritic cells pulsed withHERV-K ENV antigens exhibited phenotypes and functions con-sistent with a cellular immune response including T-cell prolif-eration, IFNg production, and HERV-K–specific cytotoxic T lym-phocyte (CTL) activity. Significantly higher CTL lysis of autolo-gous tumor cells than of uninvolved normal cells was demon-strated in patients with ovarian cancer than patients with benigndiseases and further enhanced lysis was observed if T regulatorycells were depleted.

Conclusion: Endogenous retroviral gene products in ovariancancer may represent a potentially valuable new pool of tumor-associated antigens for targeting of therapeutic vaccines to ovariancancer. Clin Cancer Res; 21(2); 471–83. �2014 AACR.

IntroductionOvarian cancer is the most common cause of death from

gynecologic malignancies and advanced ovarian cancer remainshighly lethal with >90%of patients developing tumor recurrence,resulting in 5-year survival rates of only 30% (1). Currently, thereis no test with sufficient predictive value for use in screening anddetection of premalignant or localized ovarian cancer, and70%ofpatients with ovarian cancer present with advanced disseminated

disease at the time of initial diagnosis (2). Therefore, identifica-tion of biomarkers for ovarian cancer detection, diagnosis, andimmunotherapy is imperative for improving the survival ofpatients with ovarian cancer.

Human endogenous retroviruses (HERVs) comprise approxi-mately 8% of the human genome and are believed to havegermline-integrated their DNA into the human genome over30 million years ago. HERVs exist as integrated retroviral DNAgenomes called proviruses that inevitably accumulate mutationsover time, disrupting viral genomic components and inactivatingviral activity. The most biologically active HERVs are members ofthe HERV-K superfamily, which are transcriptionally active andencode proteins that retain functionality in several human cancers(3, 4). The expression, promoter activity, and epigenetic regula-tion of HERV-Ks seem to be quite different between malignantand normal cells. However, the specific role that HERV-K proteinsplay in cancer remains an enigma. The envelope and additionalproteins of several HERVs may contribute to the development ofcancer via their fusogenic properties, and the accessory proteinsREC and NP9, alternative splicing products of the HERV-K(HML2) ENV gene, have been shown to play oncogenic functionsvia interacting with other proteins (5–7).

The central premise of successful ovarian cancer immunother-apy is that tumor cells express specific antigens that can berecognized by cytolytic T cells leading to tumor destruction(8, 9). T-cell infiltration into ovarian tumors is associated withimproved survival. In tumors with high numbers of tumor-infil-trating T cells, the expression of monokines induced by IFNg ,

1Department of Veterinary Sciences, Michale E. Keeling Center forComparative Medicine and Research, The University of Texas MDAnderson Cancer Center, Houston, Texas. 2Department of MolecularCarcinogenesis, The University of Texas MDAnderson Cancer Center,Houston, Texas. 3Graduate School of Biomedical Sciences, The Uni-versity of Texas Health Science Center at Houston, Houston, Texas.4Viral Oncology Program, SRI International, Menlo Park, California.5Department of Melanoma Medical Oncology,The University of TexasMDAnderson Cancer Center, Houston,Texas. 6Department of Immun-ology, The University of Texas MD Anderson Cancer Center, Houston,Texas. 7DivisionofGynecologicOncologyandLyndonBainesJohnsonHospital, The University of Texas MD Anderson Cancer Center, Hous-ton, Texas.

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Corresponding Author: Feng Wang-Johanning, SRI International, 333 Ravens-wood Avenue, Menlo Park, CA 94025-3493. Phone: 650-859-3271; Fax: 650-859-3153; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-14-0388

�2014 American Association for Cancer Research.

ClinicalCancerResearch

www.aacrjournals.org 471

on December 20, 2020. © 2015 American Association for Cancer Research.clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst November 4, 2014; DOI: 10.1158/1078-0432.CCR-14-0388

http://clincancerres.aacrjournals.org/

macrophage-derived chemokines, and secondary lymphoid–tis-sue chemokines are significantly increased as compared withtumors lacking T cells (10, 11). Strong positive correlations havealsobeenobserved between levels ofCD8þT cells and granzymeBwithin ovarian tumors, indicating that the majority of CD8þ Tcells are cytotoxic (12). Ovarian cancer s are immunogenic (13),suggesting that immunotherapy strategies utilizing these cytotox-ic T cells might prove effective if an appropriate tumor-associatedantigen (TAA) could be identified. However, a major obstacle inthe current development of ovarian cancer vaccines is the lack ofclearly definedTAAs that are capable of being recognized by T cells(i.e., T-cell epitopes).

Overcoming suppressive mechanisms in the tumor microen-vironment to enhance efficacy is another major challenge forcurrent approaches to ovarian cancer vaccination. Various studieshave shown that ovarian cancers escape immune surveillancewithhigh efficiency via creating a tolerogenic microenvironment (14).The tumor microenvironment mediates induction of an immu-nosuppressive programmed cell death (PD-1) pathway, with PD-1 functioning as an inhibitory surface receptor expressedbyT cells,B cells, natural killer T cells, monocytes, and dendritic cells (DC).Tumors can exploit the PD-1 inhibitory pathway to silence theimmune system (15), and expression of this molecule is inverselycorrelated with survival of patients with ovarian cancer (16).Along with myeloid cells, regulatory T cells (Tregs) within thetumor microenvironment are a crucial component of the tumorimmunosuppressive network. These cells are a heterogeneousCD4þ T-cell subpopulation that can be divided into two subsets:naturally occurring Tregs with CD4þCD25þFOXP3þ phenotypeand induced Tregs with variable CD25 expression (17). Tregsdepend on PD-1, PD-L1 (programmed cell death ligand 1), orCTL-associated antigen 4 (CTLA-4) to carry out antitumorimmune responses (18, 19). Indeed, the presence of Tregs inovarian tumors has been associated with reduced overall survival(20–22). Specifically, CD4þCD25þFOXP3þ Treg cells correspondto poor clinical outcome in epithelial ovarian cancer, and thesecells are preferentially concentrated in the tumormass rather than

in tumor draining lymph nodes (20). In addition, the therapeuticeffect of blocking the PD-1/PD-L1 pathway was shown to corre-late with increased numbers of polyfunctional ovarian tumorantigen–specific CD8þ T cells (23). These data suggest that strat-egies aimed at Treg depletion used in combination with vaccinestimulation of effector T cells could lead to more effective treat-ments for ovarian cancer.

We have previously demonstrated that patients with breastcancer, but not normal female donors, can mount immuneresponses against expressed human endogenous retroviral ele-ments using autologousDCspulsedwith theHERV-KENV surfaceantigens (24). These findings provide strong evidence that retro-viral geneproducts are capable of acting as TAAs to activate both T-cell and B-cell responses. In addition to breast cancer, we were thefirst to report the expression of multiple HERV gene and proteinproducts in ovarian cancer cell lines and tissues (25), suggestingthat HERVsmight also be able to serve as TAAs to enable targetingof ovarian cancer. In our current study, we utilize a new patientcohort to identify targets or biomarkers that could potentially beused as diagnostic or prognostic tools in ovarian cancer or as newimmunotherapeutic targets for the disease. Most important, weexplore the development of a cancer vaccine for ovarian cancerbased on HERV-K viral ENV surface protein functioning as a TAA,using patient cells cultured from matched blood, freshly resectedtumors, adjacent uninvolved tissues, and benign tissues.

Materials and MethodsThe basic procedures for many experiments in this study have

been described previously in refs. (24–29) and in SupplementaryInformation.

Clinical samples and cell linesTissue and/or blood samples from patients with ovarian cancer

and patients with benign diseases were obtained from the MDAnderson Cancer Center (Houston, TX) according to approvedInstitutional Review Board protocols (LAB04-0083) underinformed consent. Clinicopathologic characteristics are listed inSupplementary Table S1. Normal donor blood samples (Supple-mentary Table S2)were obtained fromGulf Coast Regional BloodCenter (Houston, Texas). Ovarian cancer cell lines SKOV3,DOV13, and OVCAR3, and the immortalized human ovarianepithelial cell line, T29, were gifts from Dr. Robert C. Bast Jr.(University of TexasM.D. AndersonCancer Center, Houston, TX).

Harvesting of primary samples and preparation of tumor cellsA piece of tissue in 1 to 2 mL of cold complete NOE medium

(250 mL of Medium 199 þ 250 mL DMEM) supplemented with10% FBS, 1% penicillin–streptomycin, and 1% Glutamax wasminced into small pieces. The tissue pieces and dissociated cellswere pelleted and resuspended in Accumax solution at a concen-tration of 0.5 g/10 mL of tissue and incubated at room temper-ature for 60 minutes on a shaker. The cells were transferred to aStomacher 80 bag that was inserted into the Stomacher (SewardCo.) and run for 15 minutes at high speed. The cells were thenpassed through a 40 mm nylon mesh and separated from deadcells, debris, and red blood cells on a Ficoll-1077 gradient. Thecells were spun at 900 rpm for 5 minutes, resuspended in lowserumNOEmedia, and plated in a 6-well plate. To culture tumorspheres, 6-well plateswere coatedwith 1mLof 0.8%agarose (30),and cells were added in 2 mL per well of Mammocult media

Translational Relevance

In this study, we provide evidence that the human endog-enous retrovirus K (HERV-K) envelope (ENV) protein mayfunction as a tumor-associated antigen useful for ovariancancer detection, diagnosis, and immunotherapy. Our find-ings suggest anti–HERV-K plasma antibodies and HERV-Kreverse transcriptase activity as two potential novel biomarkersfor ovarian cancer detection and diagnosis using readily avail-able body fluids and a minimally invasive procedure. From aclinical standpoint, the availability of new biomarkers todetect ovarian cancer at an early stage is critically important,because frequently this cancer is not detected until advancedstages when treatment options are limited. In addition, HERV-K–specific T cells induced cellular immune responses towardsautologous patient ovarian cancer cells, providing evidencethat the reactivated endogenous retroviral gene products rep-resent a potentially valuable new pool of tumor-associatedantigens with great specificity for targeting of therapeuticvaccines to ovarian cancer.

Rycaj et al.

Clin Cancer Res; 21(2) January 15, 2015 Clinical Cancer Research472




supplemented with heparin and hydrocortisone. To culture asci-tes, the ascites were pelleted and live cells were separated fromdead cells, debris, and red blood cells on a Ficoll-1077 gradient.Cells were pelleted, resuspended in low serum NOE media, andplated in a 6-well plate.

Measuring reverse transcriptase activity in tumor tissues andplasma

Patient plasma was isolated using Histopaque 1077 andseparated (10 mL) on isopycnic gradients using OptiPrep(Iodixanol, AxisShield) as per the manufacturer's instructions.Primary ovarian tissues were also collected during surgery andfrozen at �80�C until processed for reverse transcriptase assay.Tissue samples were crushed in culture media using a mortarand pestle. Five microliter aliquots of fractionated plasma ortissue samples were analyzed for reverse transcriptase activityusing the EnzChek Reverse Transcriptase Assay Kit (Invitrogen,Life Technologies). The EnzChek reverse transcriptase assay kituses a PicoGreen dsDNA quantitation reagent that preferen-tially detects dsDNA or RNA-DNA heteroduplexes over single-stranded nucleic acids or free nucleotides. The reverse tran-scriptase activity in a biologic sample generates long RNA–DNAheteroduplexes from a mixture of a long poly(A) template, anoligo-dT primer, and dTTP. The RNA–DNA heteroduplexesformed are then detected by the PicoGreen reagent. A reversetranscriptase standard curve was generated using serial dilu-tions of Moloney Murine Leukemia Virus Reverse Transcriptase(MMLV-RT; Invitrogen). Fluorescence was measured with aWallac VICTOR2 plate reader (Perkin Elmer).

Preparation of dendritic cells and generation of in vitrostimulated cells

The basic procedure was depicted in Supplementary Fig. S1.Briefly, after peripheral blood mononuclear cells (PBMC) wereplated and incubated for 16 hours at 37�C, floating PBMCs wererinsed off and saved for future experiments, whereas adherentcells were incubated for 6 days with granulocyte macrophagecolony-stimulating factor (GM-CSF) and IL4 (1,000 U/mL; R&DSystems). The immature DCs were harvested and transfected withHERV-K ENV or HPV16 E6 cRNAs using FuGENE Transfectionreagent per the manufacturer's instructions (Promega). Alterna-tively, cells were transfectedwithHERV-K ENV (KSU) or E6 fusionproteins, or keyhole limpet hemocyanin (KLH) protein using theBioPORTER lipid-based transfection reagent (Genlantis). Fourhours after transfection, TNFa (1,000 U/mL) was added for anadditional 16hours at 37�C to induceDCmaturation. To generatein vitro stimulated (IVS) cells, autologous PBMCs were added toDC at a ratio of 30 to 1 on day 0. Cultures were incubated in thepresence of IL2 (10 U/mL) for 7 days to generate 1-week IVS cells,as described previously (24).

T-cell proliferation, cytokine production, and CTL assaysT-cell proliferation was evaluated by a 3H-thymidine incorpo-

ration assay. Briefly, T-cell proliferation was evaluated in patientPBMC or IVS cells by restimulation for 72 hours with DCs pulsedwith no added protein, K-SU protein, or KLH control protein, at aDC to PBMC or IVS ratio of 1:30. The remaining cells were pulsedwith 1 mCi/well of [3H]-thymidine and incubated for another 18hours at 37�C. Cells were then harvested onto filter papers,transferred to scintillation vials with scintillation fluid, and ana-lyzed on a beta counter.

Cytokine production was measured using an IFN-g enzyme-linked immunospot (ELISPOT) assay (BD Biosciences). Briefly,ELISPOT plates were coated with 10 mg/mL of purified anti-human IFN-g capture antibody and incubated for 24 hours at4�C. Plates were then blocked for 2 hours with complete mediaand PBMC or IVS cells were plated at 1 � 105 per well with DCspulsed with various HERV-K antigens plated at 5 � 103 per well.Plates were incubated for 24 hours at 37�C and then washed andincubatedwith thedetection antibodies for 2hours at 25�C. Plateswere washed and incubated with streptavidin horseradish perox-idase for 2 hours at 25�C. Plateswerewashed again anddevelopedby adding 5-bromo-4-chloro-3-indolyl-phosphate/nitro blue tet-razolium (BCIP/NBT) substrate solution. Plates were washed,allowed to dry, and spots were counted using an ELISPOT reader(C.T.L. Technologies).

CTL assays were performed in round-bottom 96-well platesusing a standard 4-hour 51Cr-release assay. Briefly, various target(primary tumor, adjacent normal tissue, or benign) cells were"pulsed" with K-SU or KLH control protein for 16 hours at 37�Cusing the BioPORTER transfection reagent. Target cells wereremoved from flasks via EDTA buffer, pelleted, and resuspendedin 500 mL media and 75 mCi of chromium. A total of 5� 104/mLtarget cells and 5 � 106/mL autologous PBMCs or IVS cells(effector cells) were resuspended inmedia. The effector and targetcells, plated at 100:1, 50:1, 25:1, and 12.5, were then combined inthe plates and incubated for 4 hours at 37�C. The plates werepelleted and supernatants (100 mL) from each well were removedand counted in a gamma counter. The % specific lysis wascalculated using the formula (experimental value � minimumlysis)/(maximum lysis � minimum lysis) � 100. Labeled targetcells in media served as the minimum lysis value and labeledtarget cells in % Triton X-100 served as the maximum value.

Results and DiscussionPatient ovarian cancer cells express HERV-K transcripts andENV protein

For many years, our laboratory has been investigating theexpression and potential functions of HERV-K in various humancancers. For example, a series of studies from our laboratory havedemonstrated that HERV-K ENV proteins are strongly expressedin invasive breast cancer and ductal carcinoma in situ in com-parison with adjacent benign tissues (27, 28), that the HERV-KENV protein can trigger antigen-specific immune responses inpatients with breast cancer (24), and that the anti–HERV-K ENVmonoclonal antibody (mAb) possesses immunotherapeuticpotential (29). We have also provided preliminary evidence forthe expression ofHERV-K ENV transcripts in ovarian cancer (25).The main goals in the current study were to further characterizethe HERV-K transcript (Supplementary Fig. S2) and proteinexpression in patient ovarian cancer cells and, more importantly,to determine whether the HERV-K ENV (K-SU) protein may beable to function as a potential TAA to elicit immune responsesagainst ovarian cancer cells. To this end, we collected a newcohort of patients (n ¼ 89) with ovarian cancer (both epithelialand germ cell tumors) or benign ovarian diseases such as cysts(Supplementary Fig. S3; Supplementary Table S1). From thesepatients we obtained tumor (T), the matching uninvolved nor-mal (N), or benign (B) tissues, from which we also frequentlyderived single epithelial cells. Before we utilized these tissuesand/or cells for HERV-K studies, we characterized the phenotypic

HERV-K Expression and Immunity in Ovarian Cancer Cells

www.aacrjournals.org Clin Cancer Res; 21(2) January 15, 2015 473




and certain functional properties of epithelial ovarian cancercells. The results revealed that both primary and metastatic(ascites) ovarian cancers manifested certain phenotypes (i.e.,marker expression profiles) as well as properties (e.g., formingspheroids) of cancer stem cells (CSC) and circulating tumor cells(see Supplementary Results, Supplementary Table S3, and Sup-plementary Figs. S4 and S5). Considering the accumulatingevidence for the importance of CSCs in therapy resistance andtumor relapse and of circulating tumor cells in disseminationand metastasis (31–40), our characterizations provide potentialclues to why ovarian cancer cells are generally very aggressive,invasive, drug-resistant, and metastatic.

Subsequently, we thoroughly analyzedmRNA levels in ovariancancer cells of seven HERV transcripts, i.e., ERV3, HERV-E4-1,HERV-K type 1 (KTY1), HERV-K type 2 (KTY2), NP9 (or REC),GAG, and P1/P3 (Supplementary Fig. S2) using specific reverse

transcriptase (RT)-PCR primers (Supplementary Table S4). Thefirst two transcripts analyzed (i.e., ERV3 and HERV-E4-1) werespecific for theENV genes of g-retroviruses, whereas the remainder5 transcripts were specific toHERV-K (Supplementary Fig. S2).Wedetected the expression of most of these transcripts in bothprimary tumors (e.g., 164T, 104T, and 153T; Fig. 1B) and ascitessamples (e.g., Acc65 an Acc222; Fig. 1C). Of interest, tumorsamples seemed to generally express higher levels of many ofthese transcripts than the correspondingbenign (B) or uninvolvednormal (N) tissues (Fig. 1A). Indeed, densitometric quantificationrevealed higher Kty1 and Kty2mRNA levels in tumors than in B/Ntissues (Fig. 1A, right). The ascites samples expressed high mRNAlevels of these transcripts, which further increased slightly inspheres (Fig. 1C).

Subsequently, using the 6H5mAb(29),we analyzed theHERV-K ENVprotein expression inmultiple cultured ovarian cancer cells

A

C

E

mIgG 6H5

Acc

105-

TA

cc10

5-N

D

F G

B

E4-1

NTCβ-A

ctin

KTy1KTy

2ERV3

NP9

gag

P1/P3

1 kb

M

116-B 143-B

E4-1

NTCβ-A

ctin

KTy1KTy

2ERV3

NP9

gag

P1/P3

1 kb

M

153-T104-T

DOV13MCF10A

10.76% 62.1%

OVCAR3 SKOV3

87.35% 92.76%

Cou

nt

Acc222mIgG

6H5

Acc151-A

89%

Acc156-T

57.7%

Acc158-T

28.8%

Cou

nt

Acc65-A

65.5%

Acc

222

asci

tes

Acc

222

sphe

res

NTC

E4-1

ERV3

gag

NP9KTy2

KTy1

Mar

ker

P1/P

3

Asc

ites

NTC

E4-1

ERV3

NP9KTy2

KTy1

β-Ac

tin

Mar

ker

1 kb750 bp

500 bp

1.5 kb2 kb

Sph

eres

1 kb

500 bp

750 bp1.5 kb

2 kb

Acc65

E4-1

NTCβ-A

ctin

KTy1

KTy2ERV3

NP9

gag

P1/P3

1 kb

M

164-T

164-N

0

10

20

30

40

50

Rel

ativ

e m

RN

A le

vels

Tumor(n = 14)

B/N(n = 11)

KTy1KTy2

* *

β-Ac

tin

Figure 1.Ovarian cancer cells express HERV transcripts and HERV-K ENV protein. A, RT-PCR analysis of 7 HERV transcripts in the matching tumor (T) and uninvolved normal(N) tissues. Shown on the left is a representative RT experiment in Acc164 T/N. RT-PCR of no temple control (NTC) and b-actin was used as negative andpositive controls, respectively. Shown on the right is densitometric quantification of Kty1 and Kty2 mRNA levels in tumor versus B/N samples. Values represent(Kty1 or Kty2 OD � corresponding b-actin OD) � 100. � , P < 0.05 compared with the B/N. B, RT-PCR analysis of HERV transcripts in two benign (Acc116andAcc143) and two tumor (Acc104andAcc153) samples. Arrowheads point to the 1,000-bpmarker. C, RT-PCR analysis ofHERV transcripts in ascites and spheres ofAcc222 (left) and Acc65 (right) samples. Molecular ladder was indicated on the right and two arrowheads (left) point to the 1,000-bp marker. D, quantitation ofHERV-K ENV molecules on the surface of ovarian epithelial and cancer cells via QIFI KIT and FCM analysis. The gray line represents the isotype control, red lineHERV-K positive population, and blue line a series of 6 beadpopulations coatedwith definedquantities of amousemAb (high-affinity anti-humanCD5, cloneCRIS-1).The percentages of HERV-K–positive cells are indicated above and the numbers of HERV-K surface molecules in the 4 cell types (i.e., MCF-10A, DOV13,OVCAR3, and SKOV3) are 39,759, 1.86 � 108, 4.9 � 108, and 3.61 � 108, respectively. E, quantitation of HERV-K ENV molecules on the surface of two each ovariancancer patient primary tumors (T) and ascites (A). F and G, detection of surface expression of HERV-K ENV protein in primary Acc105 (F) and Acc222 (G)tumor cells derived from ascites by immunofluorescence and fluorescence microscopy using the mAb 6H5. Blue (DAPI) represents a nuclear stain and green (Alexafluor 488) represents surface HERV-K expression. Note that the uninvolved (N) ovarian epithelial cells from Acc105 were negative for HERV-K ENVprotein (F). mIgG, mouse IgG.

Rycaj et al.





as well as patient ovarian cancer samples (Fig. 1D–G; Supple-mentary Figs. S4B and S4D, and S5C–S5H; Supplementary TableS3). We first utilized flow cytometry (FCM) and cultured ovariancancer cells (DOV13, OVCAR3, and SKOV3) to perform quanti-tative indirect immunofluorescence (QIFI) assay (see Supplemen-tary Methods) to determine the number of HERV-K ENV mole-cules on the ovarian cancer cell surface (Fig. 1D and Supplemen-tary Fig. S5C). Using the titrated optimal antibody concentration,FCM analysis revealed that cultured ovarian cancer cells expressedmuch higher levels of ENV protein than immortalized but non-tumorigenic human ovarian epithelial cells T29 (SupplementaryFig. S5D) or the immortalized, nontransformed human mam-mary epithelial cells MCF-10A (Fig. 1D). Significantly, FCManalysis detected HERV-K surface expression in primary ovariancancer cells as well as ascites (Fig. 1E and Supplementary Fig. S4Band S4D; Supplementary Table S3). As observedwith theHERV-Ktranscript mRNA expression patterns, ascites samples generallyexpressed high levels of ENV protein in comparison with primarytumor or benign samples (Fig. 1E and Supplementary Figs. S4Band S4D, and S5E; Supplementary Table S3). Also, sphere culturesseemed to further enrich for HERV-K ENV–positive cells (Sup-plementary Fig. S4B and S4D).

Indirect immunofluorescence microscopy confirmed theexpression of HERV-K ENV protein on the surface of patientovarian cancer ascites (Fig. 1F and G and Supplementary Fig.S5F) and cultured DOV13 (Supplementary Fig. S5G) cells but notuninvolved normal ovarian epithelial cells (i.e., Acc105-N; Fig.1F). Interestingly, permeabilized DOV13 cells showed an intra-cellular pool of the ENV protein (Supplementary Fig. S5G),suggesting that not all expressed ENV protein is targeted to thecell surface. Immunohistochemistry using the 6H5 mAb alsodetected positive HERV-K ENV protein expression in both endo-metrioid and serous adenocarcinomas but not in benign cyst ornormal ovarian tissues (Supplementary Fig. S5H). Finally, wedetected, on Western blotting analysis, the HERV-K ENV proteinin fractionated plasma samples from patients with ovarian cancerbut not fromnormal donors (Fig. 2A, see below; data not shown).

Taken together, data shown here provides evidence for theexpression of various HERV transcripts and HERV-K ENV proteinin ovarian cancer patient tumors and ascites.

Ovarian cancer patient tumors and plasma possess reversetranscriptase activity and ovarian cancer patient sera containHERV-K ENV immunoreactive antibodies

Current diagnosticmethods for cancer, especiallymalignanciesof female reproductive tract, mainly depend on typical symptomsand screening tests. However, as is the case with ovarian cancer,many tumors are asymptomatic at early stage, and screening testsutilizing specific biomarkers are not sensitive enough or suffi-ciently specific for tumor detection. Therefore, simple and accu-rate tests to detect early-stage breast cancer and ovarian cancer arelacking. Recent data fromour group suggest that detection of anti-HERV antibodies in patient sera might be a useful method forrapidly screening women for breast cancer. We have shown thecorrelation between anti–HERV-K antibody titers and the type ofproliferative lesion in human breast cancer serum samples, sug-gesting that the presence and amount of anti–HERV-K antibodiesmight have a diagnostic value for breast cancer (41). In fact,HERV-K serum antibody titers showed greater sensitivity andspecificity than other recently proposed serum biomarkers ofbreast cancer (41).

In our current study, we explored the possibility that theanti–HERV-K antibodies might also be present in the ovariancancer patient sera and their titers might have some diagnosticvalues for ovarian cancer. The P1/P3 primers used in our RT-PCR analyses detect a approximately 1.6 kb transcript spanningboth POL and ENV genes (Supplementary Fig. S2). Detection ofthe P1/P3 product in many of the ovarian cancer samples (Fig.1A–C) suggests that ovarian cancer cells may also express the polgene product reverse transcriptase. To test this possibility, wecollected blood samples from cohorts of normal female donors(ND), patients with benign ovarian diseases such as cysts andcystadenomas, and patients with ovarian cancer and used thefractionated plasma samples to perform Western blottingexperiments (Fig. 2A) and to measure reverse transcriptaseactivity (Fig. 2B). The plasma fractionation approach was usedbecause we were interested in identifying where the retroviruseswere actually located in the plasma, and we found that retro-viruses banded at a density of approximately 1.16 g/mL inisopycnic density gradients. This density corresponded to aposition between gradient fraction pools A and B describedbelow. We prepared a total of 20 isopycnic fractions (i.e., from1 to 20), the first 8 fractions of which were used in Westernblotting. As shown in Fig. 2A, the HERV-K ENV protein wasdetected in the density gradient-generated fractions of plasmasamples from patients with ovarian cancer but much less frombenign patients and not from the ND. To measure the reversetranscriptase activity, we combined fractions 1–5 into "pooled"fraction A, 6–10 into fraction B, 11–15 into fraction C, and 16–20 into Fraction D. As shown in Fig. 2B, plasma samples frompatients with ovarian cancer in all fractions showed higherlevels of reverse transcriptase activity than those from the ND(also see Supplementary Results) although the plasma HERV-Kreverse transcriptase activities were not significantly differentbetween the patients with ovarian cancer and the patients withbenign diseases. Interestingly, plasma samples from patientswith benign diseases also showed higher levels of reversetranscriptase activity than those from the ND although theovarian cancer and benign plasma samples did not manifestany difference in reverse transcriptase activity (Fig. 2B).

Receiver operating characteristics (ROC) curves to distinguishbetween tumor, benign, and normal samples were used to assessthe diagnostic accuracy of our reverse transcriptase activity assays.Ideally, a reverse transcriptase activity assay should effectivelydistinguish between a control and an ovarian cancer patient usingreverse transcriptase activity as a cutoff that is not generallyreached in healthy women or women with benign diseases. Thearea under the (ROC) curve (AUC) was calculated to determinethe accuracy of a diagnostic test and the AUC scores had thefollowing specifications: 0.90–1.0¼ excellent; 0.80–0.90¼ good;0.70–0.80 ¼ fair; 0.60–0.70 ¼ poor; and 0.50–0.60 ¼ fail. Aspresented in Fig. 2B (table on the right), all diagnostic reversetranscriptase activity screens comparing benign and normal sam-ples were considered "good" regarding accuracy for all plasmafractions and almost all diagnostic reverse transcriptase activityscreens comparing normal and tumor samples were "good." Wealsomeasured the reverse transcriptase activity in lysates preparedfrom ovarian cancer, adjacent uninvolved (normal) tissues, andbenign lesions. The results revealed significantly higher reversetranscriptase activity in tumors than either normal or benigntissues (Fig. 2C; also see Supplementary Results). Diagnosticreverse transcriptase activity screens comparing benign and tumor






as well as comparing normal and tumor samples were considered"excellent" regarding accuracy. Finally, we employed the recom-binant HERV-K ENV protein in ELISA assays (see SupplementaryMethods) to determine whether the ovarian cancer patient serumcontained anti–HERV-K antibodies (Fig. 3). The results revealedhigher anti–HERV-K ENV antibody titers in patients with ovariancancer than in benign or normal female controls (Fig. 3A and B).Of potential interest, the titers in the sera of patient Acc14decreased 12 months after surgery (Fig. 3A).

Combined, data presented herein indicate that the ovariancancer cells in patient tumors also express readily detectable andhigh reverse transcriptase activity, suggesting that the HERV-K polgene, like the ENV gene, is expressed into protein. The fact that wecan detect the reverse transcriptase activity in the plasma ofpatients with ovarian cancer and patients with benign ovariandiseases suggests that either HERV-K–expressing ovarian cellsand/or HERV-K viral particles are shed into the blood circulation.The seeming discrepancy between the high reverse transcriptase

Comparison

B vs. T: Fraction A

AUC

Yes

t test P value

7.49E–06

3.25E–07

0.5923

Good

0.8890

0.8356

ROC curveSignificance

N vs. T: Fraction A

B vs. N: Fraction A

B vs. T: Fraction B

N vs. T: Fraction B

B vs. N: Fraction B

Yes Good

Yes

0.5152

0.7967

0.8839

B vs. T: Fraction C

N vs. T: Fraction C

B vs. N: Fraction C

B vs. T: Fraction D

N vs. T: Fraction D

B vs. N: Fraction D

3.09E–01

3.26E–05

2.68E–01

4.91E–08

4.47E–01

2.66E–04

4.09E–06

8.90E–01

1.07E–03

7.77E–06

Yes

Yes

Yes

0.5197

0.7200

0.8348

0.5349

0.7011

0.8219

Yes

Yes

Good

Good

Good

GoodA B C D A B C D A B C D

Normal (n = 25) Benign (n = 31) Tumor (n = 36)

A

B

Fraction # 1 2 3 4 5 6 7 8 Env

ND004 (normal donor) Acc195 (serous cystadenoma) Acc204 (serous adenocarcinoma)

1.23 1.18 1.15 1.13 1.12 1.10 1.09 1.08

1 2 3 4 5 6 7 8 Env

Density 1.23 1.18 1.15 1.13 1.12 1.10 1.09 1.08 1.23 1.18 1.15 1.13 1.12 1.10 1.09 1.08

50 kD

250 kD150 kD100 kD

75 kD

37 kD

20 kD25 kD

1 2 3 4 5 6 7 8 Env

C

2.0

1.5

1.0

0.5

RT

act

ivity

(nor

mal

ized

uni

t)

0.0

2.5

Fraction

0

0.02

0.04

0.06

0.08

0.1

Tumor Benign Normal

Comparison

B vs. T

AUC

Yes

t test P value

0.0031

0.5068

0.931

Excellent

0.5

0.006

1

ROC curve

B vs. N

N vs. T

Significance

Yes Excellent

153T

151T

158T

141T

191T

104T 69

T92

T11

2T17

8T10

5T13

9T16

1T16

4T11

8T14

0B18

7B14

4B11

1B17

6B15

0B14

3B 88B11

6B 68B10

2B16

5B14

8B18

6B13

9N11

1N11

2N15

8N13

9N10

5N

RT

act

ivity

(nor

mal

ized

uni

t)

Tumor Benign and Normal0.00

0.02

0.04

0.06

0.08

0.10

(n = 15) (n = 20)

P = 0.007

RT

act

ivity

(nor

mal

ized

uni

t)

Figure 2.Detection of HERV-K protein and reverse transcriptase (RT) activity in patient plasma. A, detection of HERV-K ENV protein in plasma fractions by immunoblot using6H5 mAb. The recombinant HERV-K ENV protein (Env) was used as a positive control. The density gradients and fraction numbers are indicated below. Notethat the HERV-K ENV proteins in Acc204 plasma fractions ran slightly slower than the recombinant Env probably due to protein glycosylations in patient tumors. B,reverse transcriptase activity was compared in pooled plasma fractions (see text for descriptions) from normal female donors or patients with ovariancancer or benign lesions. Statistical data are presented in the table on the right. A guide for classifying the accuracy of a diagnostic test is the traditional academicpoint system with the following specifications: 0.90–1.0 ¼ excellent; 0.80–0.90 ¼ good; 0.70–0.80 ¼ fair; 0.60–0.70 ¼ poor; 0.50–0.60 ¼ fail. C, reversetranscriptase activity was compared in tissues from patients with ovarian cancer (tumor) or adjacent normal tissue (normal) or from benign ovarian diseases(benign). Statistical data are displayed in the table (see the text for description).

Rycaj et al.





activity in benign patient plasma (Fig. 2B) versus very low reversetranscriptase activity from the benign tissues (Fig. 2C) couldpotentially be explained by the possibility that patients withbenign ovarian conditions also shed HERV-K–positive cells intothe circulation from other diseased tissues. Our observations thatthe sera in patients with ovarian cancer contain HERV-K immu-noreactive antibodies are interesting and raise the possibility thattheHERV-K expressed in the ovarian cancer cells can actually elicita humoral immune response against the autologous ovariancancer cells. These findings suggest HERV-K reverse transcriptaseactivity and anti–HERV-K plasma antibodies as two potentialnovel biomarkers for ovarian cancer detection and diagnosis. Theanalysis of serum for HERV-K antibodies may enable clinicians tomore easily identifywomenwithnoninvasive early ovarian cancerand treat these women at an early stage before the ovarian cancercan spread to other tissues.

HERV-K ENV protein can elicit antigen-specific T-cellproliferation and IFNg production

Consistent with the abovementioned possibility that HERV-KENV protein in ovarian cancer cells could elicit a humoral

immune response, we have previously demonstrated that theHERV-K ENVprotein can trigger cell-mediated immune responsesin patientswith breast cancer (24). In subsequent experiments, weasked whether the HERV-K ENV (KSU) protein could activate Tcells prepared from thepatientswithovarian cancer (Figs. 4 and5)and could function as TAAs to elicit T-cell–specific immuneresponses against the autologous ovarian cancer cells (Fig. 6, Fig.7). Basic procedures in preparing peripheral blood mononuclearcells (PBMC) and various immune cells including T cells, DCs,and in vitro sensitized (IVS) cells (i.e., the T cells activated byHERV-K primed DCs) were depicted in Supplementary Fig. S1.

A profound biologic issue in limiting the efficacy of cancervaccines is the failure of T cells to expand in response to antigenicstimulation. Therefore, we first analyzed the proliferativeresponses, by H3-thymidine incorporation assays, of PBMCs(mainly T cells with adherent DCs depleted; Supplementary Fig.S1) and the corresponding IVS T cells, prepared from 3 patientswith ovarian cancer and 3 ND, to either unprimed DCs or DCsprimedwithHERV-K ENV cRNAorHPV16 E6 cRNA (Fig. 4A). Theresults showed that the ovarian cancer patient PBMCs exhibitedrobust proliferative response to the autologous DCs primed withHERV-K ENV cRNA compared with unprimed DCs as well as E6-primed DCs (Fig. 4A). Importantly, the IVS cells showed furtherincreased proliferative response to the HERV-K pulsed DCs andthe HERV-K–specific proliferative responses were stronger inpatients with ovarian cancer compared with ND (Fig. 4A). Tofurther corroborate these findings, we primed DCs with therecombinant KSU protein and carried out similar thymidineincorporation assays. As illustrated in Fig. 4B, the PBMCs andIVS cells from an adenocarcinoma patient (Acc72) showed higherproliferative responses to KSU-pulsed DCs compared with thosefrom two benign samples (i.e., Acc70 and Acc91). It should benoted that PBMCs and/or IVS cells in some samples (such asAcc14 and 15 in Fig. 4A and Acc72 in Fig. 4B) also displayed someproliferative response to HERV-K ENV cRNA or protein pulsedDCs, which was not particularly surprising considering wide-spread HPV infection in the female population. Therefore, weperformed yet another experiment in which we utilized DCspulsed with the recombinant KLH protein as the control (Sup-plementary Fig. S1). As shown in Fig. 4C, the IVS cells fromahigh-grade serous adenocarcinoma patient (Acc22; SupplementaryTable S1) showed higher proliferative responses to KSU-pulsedDCs comparedwith those fromone normal donor (i.e., ND20) orfrom two benign samples (i.e., Acc16 and Acc21; SupplementaryTable S1). In this experiment, the KLH-pulsedDCs elicited similarresponses in Acc22 IVS cells to the unprimedDCs (Fig. 4C).Whenwe analyzed the pooled data of proliferative responses, ovariancancer patient IVS cells showed significantly higher proliferationrates when combined with DCs primed with HERV-K than withDCs primedwith KLH (Fig. 4D). Importantly, IVS cells in patientswith ovarian cancer showed significantly higher proliferationrates when combined with DCs primed with KERV-K than inpatients with benign diseases (Fig. 4D).

IFNg is secreted by activated CD4þ and CD8þ cells and iscritical for innate and adaptive immunity and for tumor con-trol. Therefore, we assessed HERV-K–specific T-cell responses bymeasuring IFNg production using the IFNg ELISPOT (Fig. 5; seeMaterials and Methods for experimental details). When the IVScells but not PBMCs from ovarian cancer patient Acc103 (amoderately to poorly differentiated adenocarcinoma; Supple-mentary Table S1) were incubated with KSU-pulsed DCs (i.e.,

Figure 3.Detection of anti–HERV-K antibodies in ovarian cancer patient serum. A,detection of anti–HERV-K antibodies in serially diluted serum samplesobtained from 2 patients with ovarian cancer (Acc14 and Acc22) and 4patients with benign ovarian diseases (Acc16, 19, 32 and 34). 0 moindicates that serum was drawn before the surgery while 6 mo and 12 moindicate serum draws 6 or 12 months after surgery, respectively. The anti–HERV-K antibody titers were presented as OD readings at 405 nm byELISA. B, summary of the ELISA results of anti–HERV-K antibodies in serumsamples from patients with ovarian cancer (n¼ 24), and patients with benigndiseases and normal donors (n ¼ 20). P value was based on Student t test.






DCþKSU), a large number of IFNg-producing T-cell cloneswere observed (Fig. 5A). This effect was much less obvious inthe same IVS cells stimulated with unprimed DCs or DCspulsed with KLH (Fig. 5A). Strikingly, IVS cells (and PBMCs)prepared from 3 ND did not show any IFNg response to KSU-primed DCs (Fig. 5A). Similarly, IVS cells prepared fromAcc153 (a high-grade papillary serous carcinoma; Supplemen-tary Table S1) blood samples collected at the time of surgery or6 months after surgery (Fig. 5B), or from Acc156 (LMP) andAcc158 (a serous carcinoma; Fig. 5C) also demonstrated robustIFNg response to KSU-primed DCs. Interestingly, the HERV-Ktransmembrane (KTM) protein, which has been reported tohave immunosuppressive properties (42), demonstrated sim-ilar IFNg-stimulating effects to KSU in activating DCs in these 3patients with ovarian cancer (Fig. 5B and C) but not in a ND-derived DCs (ND25; Fig. 5B). Again, when the IVS cells fromthe same patients were combined with unprimed DCs or DCspulsed with KLH protein, minimal IFNg response was observed(Fig. 5B and C). Finally, when we analyzed the pooled data ofIFNg response in KSU-pulsed DCs from patients with ovarian

cancer (tKSU) versus those from benign patients (bKSU), weobserved significantly more IFNg-positive spots in IVS cellscoincubated with tKSU DCs than bKSU DCs (P ¼ 0.0061; Fig.5D).

Altogether, these results (Fig. 4 and 5) suggest that thehuman HERV-K gene product, i.e., the ENV protein, is immu-nogenic and can elicit notable T-cell responses measured by T-cell proliferation and IFNg production. Of great interest, theKSU-triggered T-cell responses are much stronger in ovariancancer samples than in benign patients and normal femaledonors, presumably because the T cells in patients with ovariancancer are already partially stimulated by the endogenous DCs,which are partially primed by the reactivated HERV-Kexpression.

HERV-K–specific T-cell cytotoxicity in ovarian cancer patientsOne of the main goals of immunotherapy is to generate

cytotoxic T lymphocytes (CTL) that can recognize antigens thatare specifically expressed on the tumors, thereby leading to tumordestruction. However, there is considerablemolecular diversity in

C

B A

PBMC

IVS

PBMC

IVS

PBMC

IVS

PBMC

IVS

PBMC

IVS

PBMC

IVS

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

CP

M

Media only

DC only

DC+K cRNA

DC+E6 cRNA

Acc7 Acc14 Acc15 ND06 ND10 ND20PBM

CIV

S

PBMC

IVS

PBMC

IVS

0

200

400

600

800

1,000

1,200

1,400

CP

M

Media only

DC only

DC+KSU

DC+E6

Acc70 Acc72 Acc91

0

200

400

600

800

Media only

DC

DC+KSU

DC+KLH

CP

M

Acc16 ND20Acc22Acc21

D

CP

M

bKLH

IVS

(n

= 7

)

bKSU IV

S

(n

= 7

)tK

LH IV

S

(n

= 5

)

tKSU IV

S

(n

= 8

)

0

5,000

10,000

15,000P = 0.007

P = 0.011

Figure 4.HERV-KENV induced T-cell proliferation. A, H3-thymidine incorporation assays in PBMCs and IVS cells prepared from3patientswith ovarian cancer (Acc7, 14, and 15)and from 3 normal female donors (ND). The PBMCs and IVS cells (Supplementary Fig. S4) were incubated with media alone, unprimed autologous DCs, orautologous DCs primed with either HERV-K ENV or HPV16 E6 (control) cRNA for 72 hours in the presence of H3-thymidine. Shown are the CPM (counts per minutes;mean � S.D). Note that for unknown reasons lower CPM readings were observed in the DC wells. B and C, H3-thymidine incorporation assays performedsimilarly to those described above in A except autologous DCs were pulsed with recombinant KSU or E6 protein (B) or KLH protein (C). See text for more details.Note that the protein-primed DCs (B and C) generally elicited lower T-cell proliferative responses than the cRNA-pulsed DCs (A), presumably due to well-known lower efficiency in protein transduction compared to RNA/DNA transfection. D, analysis of pooled data of proliferative responses. See text for detaileddescriptions.

Rycaj et al.





ovarian cancer, and a major factor limiting vaccine progress is thelack of well-characterized rejection antigens (13). IFNg produc-tion and proliferation are clear signs of specific activation; how-ever, they donot prove cytotoxic ability.Our paramount objectivewas to determine the ability of antigen-specific T cells to killautologous target cells expressing HERV-K while ignoring benignor normal cells. Our cohort of samples provided us a uniqueopportunity to achieve this objective in that it allowed cytolyticactivity of autologous target cells (instead of an established cancerline not specific to the patient) to be assessed. To this end, weprepared KSU-primed IVS cells from ovarian cancer patientAcc115 (a malignant mixed mullerian tumor; SupplementaryTable S1) and autologous tumor cells and, for comparison,normal cells from the adjacent uninvolved tissues (Fig. 6A). Usingthe IVS cells as the "effector" cells and autologous tumor ornormal cells as "target" cells in a CTL (i.e., 51Cr release) assay,we observed higher specific target cell lysis in tumor cells than thenormal cells (Fig. 6A). Significantly, the % specific lysis wasincreased dramatically in tumor cells when the target cells werealso loadedwith the KSU (Fig. 6A). Similar CTL assays in a total of6 patients with ovarian cancer revealed that percentage of CTLlysis of autologous tumor cells expressing KSU antigen wassignificantly higher than that of uninvolved normal cells (Fig.

6B). Similarly, significantly higher % of target cell killing wasobserved with tumor cells prepared from28 patients with ovariancancer than with cells from 13 patients with benign ovariandiseases (Fig. 6C). These results demonstrate the efficacy ofHERV-K–specific T cells in preferentially killing autologousHERV-K–expressing ovarian cancer cells. The results also suggestthat PBMCs in circulation from ovarian cancer patients containHERV-K–specific CTLs that can be restimulated in vitro toinduce cytolytic activity against HERV-K–expressing ovariantumor cells.

Then why are ovarian cancer cells in patient tumors not elim-inated by the HERV K-primed CTLs? One explanation may be theoverall immunosuppressive tumor microenvironment that is lim-iting the functions of antigen-specific CTLs (8, 14, 17, 19, 42). Asmentioned in the section Introduction, Tregs represent a subpop-ulation of immunosuppressive T lymphocytes that dampen theimmune response and accumulation of intratumoral Tregs hasbeen associated with a high mortality rate in patients with ovariancancer. Likely, it is the intricate balance between the abundance ofeffector T cells and that of Tregs that determines patient outcome.In fact, some studies have shown that ovarian cancer cells escapeimmune surveillance via creationof an immunosuppressivemicro-environment (14) and inductionof suppressive Tregs is frequent in

Figure 5.HERV-K ENV induced IFNg production. A, ELISPOT assay for IFNg-producing T-cell clones. HERV-K SU (KSU) IVS or PBMCs from Acc103 ovarian cancer patient orfrom 3 ND were combined with DCs pulsed with HERV-K protein (DCþKSU), DCs pulsed with KLH control protein (DCþKLH), media alone, or unpulsed DCs.An image of an ELISPOT plate is displayed and a purple spot represents one IFNg–secreting cell. B and C, ELISPOT assays in more ovarian cancer patients.Experiments were conducted as in A except with an additional condition in which DCs pulsed with the HERV-K transmembrane protein (KTM) were also used. D,summary of IFNg ELISPOT data for patients with ovarian cancer (all t samples; n¼ 10) compared with patients with benign disease (all b samples; n¼ 10). Note thenegligible IFNg responses in all studies with PBMCs.






ovarian cancer tumor lesions (14). We first characterized thecomposition of cellular populations in our IVS cells by stainingwith antibodies against CD4 (for T helper cells), CD8 (for CTLs),CD25 (for activated T and B cells), CD56 (for NK cells), and FoxP3(for Tregs). As presented in Supplementary Table S5, we observedan increase in both CD8þ and CD4þ cells in both tumor andbenign samples, an increase in activated B and T cells in all patients(except the last benign patient), and a decrease in Tregs only inovarian cancer samples.Of significance, FoxP3þ cellswere detectedin tumor-infiltrated lymphocytes in Acc213ovarian cancer (a high-grade serous tumor; Supplementary Table S5), supporting thepresence of Tregs in ovarian tumors in situ.

We subsequently determined the effect of Treg depletion (fromthe IVS effector cells) on the ability ofHERV-K specific CTLs to lyseautologous target ovarian cancer cells. As shown in Fig. 7A, inovarian cancer patients Acc104 (a high-grade serous carcinoma;Supplementary Table S1) and Acc164 (a mucinous borderlinetumor; Supplementary Table S1), target cell killing was greater intumor cells than in matching normal cells and Treg depletionslightly enhanced lysis. In 2 other patients with ovarian cancer westudied, tumor cells pulsed with KSU showed more preferentialkilling comparedwith the same cells pulsedwith KLH control andin one patient (Acc204, a serous adenocarcinoma) Treg depletion

resulted in slightly increased target cell lysis (Fig. 7B). In a cohortof 7 (unmatched) tumor and benign samples, we observed highertarget cell lysis in tumor cells than in benign cells and, in general,cell killing was enhanced when Tregs were depleted (Fig. 7C).Overall, Treg depletion (i.e., Treg�) led to enhanced specific lysis(Fig. 7D). A comparison of lysis of tumor cells, uninvolvednormal cells, and benign cells as targets in the presence ofundepleted T cells (both), Treg�, and Tregþ cells showed prefer-ential killing of tumor cells and significantly higher percentage oflysis was observed in Treg� than Tregþ from patients with ovariancancer (n ¼ 28; P ¼ 0.0396; Fig. 7E).

In summary, in this project we have shown the expression ofHERV-K transcripts and HERV-K ENV protein in primary ovariancancer cells and ascites. In addition, the ovarian cancer tissues andpatient plasma express detectable reverse transcriptase activityand patient sera containHERV-K antibodies. Furthermore, HERV-K primed DCs can activate T-cell responses (i.e., T-cell prolifer-ation and IFNg production). Most importantly, HERV-K–specificT cells can elicit robust cytotoxicity towards autologous ovariancancer cells, which is further enhanced by Treg depletion. Theseobservations, together with the findings that HERV-K expressionappears to be reactivated in a wide spectrum of human tumors,provide rationale for employingHERV-K gene products as TAAs to

Figure 6.HERV-K–specific T-cell cytotoxicity inpatients with ovarian cancer. A, CTLassays in Acc115 ovarian cancerpatient–derived cells. Both tumor andadjacent uninvolved normal cells fromAcc115 were harvested and employedas "target cells" and the autologousIVS cells were used as "effector cells."Target cells were further pulsed withKSU protein. Effector to target cellratios ranged from 100:1 to 12.5:1. B, thepercent specific lysis from CTL assaysusing ovarian cancer target cellsversus normal (uninvolved) targetcells. Data were based on thematching samples from 6 patients.C, the percent specific lysis from CTLassays using ovarian cancer targetcells versus benign tissue-derivedtarget cells. Data were based on thenonmatching samples from 28patients with ovarian cancer and 13benign diseases.

Rycaj et al.





develop therapeutic vaccines to target ovarian cancer and othertumors.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: K. Rycaj, L.G. Radvanyi, G.L. Johanning, F. Wang-JohanningDevelopment of methodology: K. Rycaj, J.B. Plummer, B. Yin, L.G. Radvanyi,L.M. Ramondetta, F. Wang-JohanningAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): K. Rycaj, B. Yin, M. Li, J. Garza, F. Wang-Johanning

Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): K. Rycaj, J.B. Plummer, M. Li, J. Garza, L.G. Radvanyi,K. Lin, G.L. Johanning, D.G. Tang, F. Wang-JohanningWriting, review, and/or revision of the manuscript: K. Rycaj, J.B. Plummer, B.Yin,M. Li, J. Garza, L.G. Radvanyi, L.M. Ramondetta, G.L. Johanning, D.G. Tang,F. Wang-JohanningAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): K. Rycaj, J. Garza, F. Wang-JohanningStudy supervision: K. Rycaj, F. Wang-Johanning

AcknowledgmentsThe authors thank Dr. Jinsong Liu for supplying pathology reports, Jody

Folloder, Jessica Gallegos, and Donna C. Griffith for consenting patients andpatient follow-up, and Pamela Whitney for assistance in FCM. This research was

A

% S

peci

fic ly

sis

0

20

40

60

80

100

100 to 150 to 125 to 112.5 to 1

Acc204

Acc164

% S

peci

fic ly

sis

0

20

40

60

80

100 to 150 to 125 to 112.5 to 1

Tumor cells Normal cells

Treg–

Both

Treg+

Treg–

Both

Treg+

Acc104

% S

peci

fic ly

sis

0

10

20

30

40

50

100 to 150 to 125 to 112.5 to 1

Tumor cells Normal cells

Treg–

Both

Treg+

Treg–

Both

Treg+

Tumor cells+KLH Tumor cells+KSU

Treg–

Both

Treg+

Treg–

Both

Treg+

Treg–

Both

Treg+

Treg–

Both

Treg+

% S

peci

fic ly

sis

0

10

20

30

40

100 to 150 to 125 to 112.5 to 1

Acc213

Tumor cells+KSUTumor Cells+KLH

B

D

E

C

BothTreg+Treg–

P = 0.0007

P = 0.0006

% S

peci

fic ly

sis

% S

peci

fic ly

sis

–20

0

20

40

60

80

Tumor (n = 28)Benign (n = 18)Normal (n = 10)

P = 0.0076

P < 0.0001

P = 0.0027

P < 0.0001

BothTre

g–

Treg+

BothTre

g–

Treg+

BothTre

g–

Treg+

% S

peci

fic ly

sis

Treg–

Both

Treg+

Treg–

Both

Treg+

0

10

20

30

40

50

100 to 150 to 125 to 112.5 to 1

Acc102B Acc104T

% S

peci

fic ly

sis

Treg–

Both

Treg+

Treg–

Both

Treg+

-20

0

20

40

60 100 to 150 to 125 to 112.5 to 1

Acc150B Acc153T%

Spe

cific

lysi

s

Treg–

Both

Treg+

Treg–

Both

Treg+

0

20

40

60

80

100 to 150 to 125 to 112.5 to 1

Acc186B Acc191T

% S

peci

fic ly

sis

Treg–

Both

Treg+

Treg–

Both

Treg+

Treg–

Both

Treg+

Treg–

Both

Treg–

–20

0

20

40

60

80

Acc121T

100 to 150 to 125 to 112.5 to 1

Acc103T Acc71B Acc91B

% S

peci

fic ly

sis

Treg–

Both

Treg+

Treg–

Both

Treg+

Treg–

Both

Treg+

Treg+

Treg–

Both

–10

0

10

20

30

40 100 to 150 to 125 to 112.5 to 1

Acc122T Acc116B Acc111B Acc118T

80

60

40

20

0

Figure 7.Effect of Treg depletion on HERV-K–specific T-cell cytotoxicity in patients with ovarian cancer. A, tumor and adjacent normal uninvolved tissues obtained frompatient Acc104 and Acc164 were harvested and employed as "target cells." Unpurified bulk IVS cells (both) were used as "effector" cells, which were furtherseparated into purified Tregs (Tregþ) and Treg-depleted (Treg�) population. The effector to target cell ratios ranged from 100:1 to 12.5:1. B, ovarian cancer cells frompatients Acc204 and Acc213 were purified and employed as target cells for the CTL assay. IVS cells were prepared and further fractionated as in A and target cellswere further pulsed with KSU or KLH protein. C, CTL assays using a cohort of tumor (T) or benign (B) cells from the patients indicated as target cells. IVScellswere used as effector cells and target cellswerepulsedwithKSUprotein.D, the percent specific lysis fromCTLassayswas compared among adjacent uninvolvedovarian, benign ovarian, and ovarian tumor target cells (n ¼ 13). E, summary of percent specific lysis from CTL assays from normal female donors (n ¼ 10),benign disease (n¼ 18), and ovarian cancer patients (n¼ 8). Target cells were prepared from ovarian cancer or adjacent uninvolved (normal) tissues or from benigntissues. Three types of effector cells as described in A were used. The Student t test was used to calculate significance of differences between groups.






performed inpartial fulfillment of the requirements for the Ph.D. degree fromTheUniversityof TexasGraduate Schoolof Biomedical Sciences atHouston (K. Rycaj).

Grant SupportThis work was supported by a Department of Defense Ovarian Idea Devel-

opment Award (OC060086) and, in part, by Impact grant BC113114 from theU.S. Department of Defense.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received February 14, 2014; revised October 11, 2014; accepted October 13,2014; published OnlineFirst November 4, 2014.

References1. Siegel R, Ward E, Brawley O, Jemal A. Cancer statistics, 2011: the impact of

eliminating socioeconomic and racial disparities on premature cancerdeaths. CA Cancer J Clin 2011;61:212–36.

2. Rosenthal A, Jacobs I. Ovarian cancer screening. Semin Oncol 1998;25:315–25.

3. Lower R, Lower J, Kurth R. The viruses in all of us: characteristics andbiological significance of human endogenous retrovirus sequences. ProcNatl Acad Sci U S A 1996;93:5177–84.

4. Lower R, Tonjes RR, Korbmacher C, Kurth R, Lower J. Identification of aRev-related protein by analysis of spliced transcripts of the human endog-enous retroviruses HTDV/HERV-K. J Virol 1995;69:141–9.

5. Ruprecht K,Mayer J, SauterM, Roemer K,Mueller-LantzschN. Endogenousretroviruses and cancer. Cell Mol Life Sci 2008;65:3366–82.

6. Denne M, Sauter M, Armbruester V, Licht JD, Roemer K, Mueller-LantzschN. Physical and functional interactions of human endogenous retrovirusproteins Np9 and rec with the promyelocytic leukemia zinc finger protein.J Virol 2007;81:5607–16.

7. Kaufmann S, SauterM, SchmittM, Baumert B, Best B, Boese A, et al. Humanendogenous retrovirus protein Rec interacts with the testicular zinc-fingerprotein and androgen receptor. J Gen Virol 2010;91:1494–502.

8. Hwu P, Freedman RS. The immunotherapy of patients with ovarian cancer.J Immunother 2002;25:189–201.

9. Kammerer U, Thanner F, KappM, Dietl J, Sutterlin M. Expression of tumormarkers on breast and ovarian cancer cell lines. Anticancer Res 2003;23:1051–5.

10. Zhang L, Conejo-Garcia JR, KatsarosD, Gimotty PA,MassobrioM, RegnaniG, et al. Intratumoral T cells, recurrence, and survival in epithelial ovariancancer. N Engl J Med 2003;348:203–13.

11. SatoE,Olson SH,Ahn J, BundyB,NishikawaH,QianF, et al. IntraepithelialCD8þ tumor-infiltrating lymphocytes and a high CD8þ/regulatory T cellratio are associated with favorable prognosis in ovarian cancer. Proc NatlAcad Sci U S A 2005;102:18538–43.

12. Milne K, Kobel M, Kalloger SE, Barnes RO, Gao D, Gilks CB, et al.Systematic analysis of immune infiltrates in high-grade serous ovariancancer reveals CD20, FoxP3 and TIA-1 as positive prognostic factors. PLoSONE 2009;4:e6412.

13. Kandalaft LE, PowellDJ Jr, SinghN,CoukosG. Immunotherapy for ovariancancer: what's next? J Clin Oncol 2011;29:925–33.

14. Chiriva-Internati M, Mirandola L, Kast WM, Jenkins MR, Cobos E, CannonMJ. Understanding the cross-talk between ovarian tumors and immunecells: mechanisms for effective immunotherapies. Int Rev Immunol 2011;30:71–86.

15. Keir ME, Butte MJ, Freeman GJ, Sharpe AH. PD-1 and its ligands intolerance and immunity. Annu Rev Immunol 2008;26:677–704.

16. Hamanishi J, Mandai M, Iwasaki M, Okazaki T, Tanaka Y, Yamaguchi K,et al. Programmed cell death 1 ligand 1 and tumor-infiltrating CD8þ Tlymphocytes are prognostic factors of human ovarian cancer. Proc NatlAcad Sci U S A 2007;104:3360–5.

17. PrestonCC,Goode EL,Hartmann LC, Kalli KR, Knutson KL. Immunity andimmune suppression in human ovarian cancer. Immunother 2011;3:539–56.

18. Francisco LM, Salinas VH, BrownKE, Vanguri VK, FreemanGJ, KuchrooVK,et al. PD-L1 regulates the development, maintenance, and function ofinduced regulatory T cells. J Exp Med 2009;206:3015–29.

19. Zou W. Immunosuppressive networks in the tumour environment andtheir therapeutic relevance. Nat Rev Cancer 2005;5:263–74.

20. Curiel TJ, Coukos G, Zou L, Alvarez X, Cheng P, Mottram P, et al. Specificrecruitment of regulatory T cells in ovarian carcinoma fosters immuneprivilege and predicts reduced survival. Nat Med 2004;10:942–9.

21. Wolf D,Wolf AM, RumpoldH, Fiegl H, Zeimet AG,Muller-Holzner E, et al.The expression of the regulatory T cell-specific forkhead box transcriptionfactor FoxP3 is associated with poor prognosis in ovarian cancer. ClinCancer Res 2005;11:8326–31.

22. Preston CC, Maurer MJ, Oberg AL, Visscher DW, Kalli KR, Hartmann LC,et al. The ratios of CD8þ T cells to CD4þCD25þ FOXP3þ and FOXP3� Tcells correlate with poor clinical outcome in human serous ovarian cancer.PLoS ONE 2013;8:e80063.

23. Duraiswamy J, Freeman GJ, Coukos G. Therapeutic PD-1 pathwayblockade augments with other modalities of immunotherapy T-cellfunction to prevent immune decline in ovarian cancer. Cancer Res2013;73:6900–12.

24. Wang-Johanning F, Radvanyi L, Rycaj K, Plummer JB, Yan P, Sastry KJ, et al.Human endogenous retrovirus K triggers an antigen-specific immuneresponse in breast cancer patients. Cancer Res 2008;68:5869–77.

25. Wang-Johanning F, Liu J, Rycaj K, Huang M, Tsai K, Rosen DG, et al.Expression of multiple human endogenous retrovirus surface envelopeproteins in ovarian cancer. Int J Cancer 2007;120:81–90.

26. Wang-Johanning F, Gillespie GY, Grim J, Rancourt C, Alvarez RD, SiegalGP, et al. Intracellular expression of a single-chain antibody directedagainst human papillomavirus type 16 E7 oncoprotein achieves targetedantineoplastic effects. Cancer Res 1998;58:1893–900.

27. Wang-Johanning F, Frost AR, Johanning GL, Khazaeli MB, LoBuglio AF,Shaw DR, et al. Expression of human endogenous retrovirus k envelopetranscripts in human breast cancer. Clin Cancer Res 2001;7:1553–60.

28. Wang-Johanning F, Frost AR, Jian B, Epp L, Lu DW, Johanning GL.Quantitation of HERV-K env gene expression and splicing in human breastcancer. Oncogene 2003;22:1528–35.

29. Wang-Johanning F, Rycaj K, Plummer JB, Li M, Yin B, Frerich K, et al.Immunotherapeutic potential of anti-human endogenous retrovirus-Kenvelope protein antibodies in targeting breast tumors. J Natl Cancer Inst2012;104:189–210.

30. Lang JY, Hsu JL, Meric-Bernstam F, Chang CJ, Wang Q, Bao Y, et al. BikDDeliminates breast cancer initiating cells and synergizes with lapatinib forbreast cancer treatment. Cancer Cell 2011;20:341–56.

31. TangDG. Understanding cancer stem cell heterogeneity and plasticity. CellRes 2012;22:457–72.

32. Bapat SA,Mali AM, Koppikar CB, KurreyNK. Stemandprogenitor-like cellscontribute to the aggressive behavior of human epithelial ovarian cancer.Cancer Res 2005;65:3025–9.

33. Burgos-Ojeda D, Rueda BR, Buckanovich RJ. Ovarian cancer stem cellmarkers: prognostic and therapeutic implications. Cancer Lett 2012;322:1–7.

34. Ferrandina G, Bonanno G, Pierelli L, Perillo A, Procoli A, Mariotti A, et al.Expression of CD133-1 and CD133-2 in ovarian cancer. Int J GynecolCancer 2008;18:506–14.

35. Zhang S, Balch C, Chan MW, Lai HC, Matei D, Schilder JM, et al. Iden-tification and characterization of ovarian cancer-initiating cells from pri-mary human tumors. Cancer Res 2008;68:4311–20.

36. Angus B, Purvis J, Stock D, Westley BR, Samson AC, Routledge EG, et al.NCL-5D3: a new monoclonal antibody recognizing low molecular weightcytokeratins effective for immunohistochemistry using fixed paraffin-embedded tissue. J Pathol 1987;153:377–84.

37. Yu M, Bardia A, Wittner BS, Stott SL, Smas ME, Ting DT, et al. Circulatingbreast tumor cells exhibit dynamic changes in epithelial andmesenchymalcomposition. Science 2013;339:580–4.

38. RhimAD,Mirek ET, AielloNM,Maitra A, Bailey JM,McAllister F, et al., EMTand dissemination precede pancreatic tumor formation. Cell 2012;148:349–61.


Rycaj et al.




39. Pecot CV, Bischoff FZ,Mayer JA,Wong KL, Pham T, Bottsford-Miller J, et al.A novel platform for detection of CKþ and CK� CTCs. Cancer Discov2011;1:580–6.

40. Liao J, Qian F, Tchabo N, Mhawech-Fauceglia P, Beck A, Qian Z, et al.Ovarian cancer spheroid cells with stem cell-like properties contribute totumor generation, metastasis and chemotherapy resistance through hyp-oxia-resistant metabolism. PLoS ONE 2014;9:e84941.

41. Wang-Johanning F, Li M, Esteva FJ, Hess KR, Yin B, Rycaj K, et al.Human endogenous retrovirus type K antibodies and mRNA as serumbiomarkers of early-stage breast cancer. Int J Cancer 2014;134:587–595.

42. Mangeney M, Heidmann T. Tumor cells expressing a retroviral envelopeescape immune rejection in vivo. Proc Natl Acad Sci U S A 1998;95:14920–5.






2015;21:471-483. Published OnlineFirst November 4, 2014.Clin Cancer Res Kiera Rycaj, Joshua B. Plummer, Bingnan Yin, et al. toward Autologous Ovarian Cancer Cells

Specific T Cells−Cytotoxicity of Human Endogenous Retrovirus K

Updated version

10.1158/1078-0432.CCR-14-0388doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2014/11/05/1078-0432.CCR-14-0388.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/21/2/471.full#ref-list-1

This article cites 42 articles, 18 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/21/2/471.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/21/2/471To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-14-0388

http://clincancerres.aacrjournals.org/content/suppl/2014/11/05/1078-0432.CCR-14-0388.DC1

http://clincancerres.aacrjournals.org/content/21/2/471.full#ref-list-1

http://clincancerres.aacrjournals.org/content/21/2/471.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/21/2/471


cytotoxicity of human endogenous retrovirus speciﬁc t cells … · cancer cells kiera rycaj1,2,3,...

Documents